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 2006-10-25

00:00:36linuxstbThey may read the IRC logs and reply anyway, but generally it's best to hang around and ask here.
00:01:04preglowwill a splash() scroll if the string is too for the screen?
00:01:27linuxstbI thought it wrapped?
00:01:53preglowoh, it wraps
00:03:03 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
00:03:20tucozhmm. no bluebrother around :(
00:04:14 Join psiborg [0] (
00:04:27tucozbluebrother, in case you read this. The manual is broken (I think). It fails on: makeindex -s rockbox-build.idx
00:04:35tucozIndex style file not found.
00:07:08hcslinuxstb: I removed his writes to COP_CTL and CPU_CTL, I'm not getting the switching anymore, I'll be seeing over the next few hours if it still crashes...
00:07:56 Join TerrorByte [0] (i=4889bb7a@gateway/web/cgi-irc/
00:08:18TerrorByteLatest update: 'Optionally auto-unhide the next card on a stack if the last known card is moved away.'
00:08:21TerrorByteNice :)
00:10:03tucozthat does not have to be an option, but the only behaviour (imho)
00:10:35 Quit TeaSeaLancs (Read error: 145 (Connection timed out))
00:10:51 Join TeaSeaLancs [0] (n=thunderc@
00:13:18tucozbluebrother: did you forget to commit, or is that supposed to be included in standard tetex installs?
00:13:50tucozgood night
00:13:52 Part tucoz ("Leaving")
00:14:30linuxstbhcs: Probably best to ask barrywardell about it next time he's around.
00:17:33preglowlinuxstb: you got any 24 bit flac files you'd care to test with 24 bit lpc routines on h1x0?
00:17:38preglowcoldfire anyway
00:17:46 Quit breez ()
00:19:39linuxstbpreglow: I can find some on
00:20:17 Quit newbyx86_ (Read error: 110 (Connection timed out))
00:20:29preglowlinuxstb: nah, i just thought you had some around you know how sounded
00:21:27 Join newbyx86_ [0] (
00:21:30preglowjust gotta code the last part of the default lpc case and we'll see how much cpu i've saved
00:21:44linuxstbI'm not at home, so it's easier to just download some new ones. But it seems's search is broken...
00:21:45preglowat least the bloody file i've got here should be able to be played without skipping...
00:22:12linuxstbPossibly other formats like wavpack are better for 24-bit though - I don't think FLAC gets a very good compression ratio.
00:22:13preglowlinuxstb: you don't need to do my testing for me anyway :) just wondered if you already had some
00:22:19preglowlinuxstb: it doesn't
00:23:36*linuxstb wants someone to add 24-bit and varying samplerate output to the ipod audio driver.
00:24:13hcslinuxstb: I've gone ahead and made a modified patch, he can see himself
00:25:26*TerrorByte etyBrorreT
00:25:42*TerrorByte etyBrorreT
00:26:03Paul_The_NerdTerrorByte: Is there any chance asking you politely to stop will actually succeed?
00:26:12 Quit TerrorByte ("CGI:IRC")
00:26:57*Paul_The_Nerd wishes *that* were his super power.
00:27:18hcsstabbing people in the face over the internet?
00:27:33Paul_The_NerdBeing able to will people into departing.
00:27:38Paul_The_NerdI have a super power, but it's rather pathetic.
00:28:41*Paul_The_Nerd wouldn't mind 24-bit and higher sample rate audio on his iPod either.
00:28:52preglow24 bit i don't care much about
00:28:56preglowhigher sample rate would be nice, though
00:29:02preglowif only for not having to resample
00:29:16hcsI'd like the resampling to work better for 32KHz
00:29:29Paul_The_NerdResampling seems to be an inspiration for a lot of complaints about our sound quality
00:30:42preglowwell, small wonder, the resampler sucks rod
00:31:14hcs22KHz to 44KHz seems to work fine, but then again that's trivial
00:31:26preglowhcs: well, it's not
00:31:48hcspreglow: oh?
00:31:56preglowit too sounds like shit
00:31:58preglowif you listen closely
00:32:11 Join Davide-NYC [0] (
00:32:30Paul_The_NerdA sack of shit? Or was it a bag?
00:32:45Davide-NYCman! What did I just step into?
00:32:54Paul_The_NerdDavide-NYC: Fond memories.
00:33:26Paul_The_NerdSomeone once described Rockbox as sounding like such.
00:34:02Davide-NYCIs there a concensus on Simon Menzel (Rincewind) recording button patch?
00:34:13Genre9mp3Paul_The_Nerd: "Sounds like a bag of shit and looks like a piece of crap"
00:34:13 Join Febs [0] (
00:34:31 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
00:34:31preglowi might make a new resampler one day
00:34:34preglowwe'll see
00:34:46preglowbut it's not exactly a trivial job in a low-cpu environment like ours
00:35:27Davide-NYCGenre9mp3: LOL
00:36:18Genre9mp3Davide-NYC: Check first and last from here:
00:37:14Paul_The_NerdDavide-NYC: I think it'd be better if the function of the button were more hard-coded, rather than configurable.
00:38:14Davide-NYCI agree with you. Having said that I find it nearly ludicrous that the Rec button not take a user to the Recording screen from file tree menu and WPS.
00:38:27Genre9mp3Paul_The_Nerd: The point is though, what it should do?
00:38:33*Davide-NYC the 'nearly ludicrous bit will not get me on the quotes page
00:39:02Davide-NYCI find it apocalyptically block headed~!
00:39:09*Davide-NYC trying too hard
00:39:20Paul_The_NerdGenre9mp3: In the filetree, Holding it goes to the record screen, tapping it is one-button insert, Play+it is one-button Queue. In the WPS, holding it goes to the recording screen, tapping it shows you the current playlist, and Play+it is... I dunno... Playlist Catalog? EQ Preset list?
00:40:02Davide-NYCHold Rec button should go to the Recording screen.
00:40:07Genre9mp3Paul_The_Nerd: Sounds nice...
00:40:37Paul_The_NerdI think that covers what most people want to use it for.
00:40:49Genre9mp3I agree...
00:42:01Davide-NYCWhat about from Menu?
00:42:11Davide-NYCMust we exit menu to go to recording?
00:42:36Davide-NYCI say in all possible screens: hold rec = recording screen
00:42:49Genre9mp3Actually, you have to enter in the Menu to go to the recording screen ;)
00:45:05Genre9mp3But yes... I guess holding Rec could be used universally
00:45:08 Join |AhIoRoS| [0] (n=ahioros@
00:45:14Davide-NYCYay! Lets do it!
00:45:36*Davide-NYC waits for a dev to save us
00:45:48preglowwell, unless some better use for it comes up
00:45:50preglowi say that's a good idea
00:46:11Davide-NYC(in my best Cali voice) It's been like, forever.
00:46:19preglowit is a record button, after all
00:46:44Paul_The_NerdI think long hold of record should *definitely* be the record screen, pretty much everywhere (except places where quick-exiting doesn't make sense for some reason, if those exist)
00:47:15preglowi agree
00:47:18Genre9mp3Should be tap = insert and play+Rec = queue or vice versa?
00:47:34Genre9mp3Also... taping rec in the menu doesn't get you back?
00:47:41Paul_The_NerdI can't imagine a better use for long-hold, since any real 'feature' we'd want to put on a button universally available, and recording is only available on recording targets anyway.
00:48:10Davide-NYCGenre9mp3: On H1x0, no
00:48:11jhMikeSdon't forget the FMRadio screen. There's no use of long rec there.
00:48:27Davide-NYCThat should go to Rec with FM enabled and chosen as source
00:48:40Paul_The_NerdGenre9mp3: Actually, tapping should be queue, holding should be insert, just so you can't accidentally permanently change playlists very easily.
00:48:49jhMikeSAlready does, just need to do it with the key instead of the menu
00:49:03Paul_The_NerdI can imagine tapping it, skipping over the song once you realize you've done so, and then saving it later and being irritated to find it was in there.
00:49:06Davide-NYCright, I was just being explicit
00:49:27Davide-NYCI can;t believe there's no feature request for this
00:49:30Genre9mp3Paul_The_Nerd: Yes... so it's better the opposite from your very first proposition
00:49:33Davide-NYCIf there is I can't find it
00:49:54Paul_The_NerdGenre9mp3: The original proposition was based partially on *my* usage habits, but I realized that swapping them is better for the general public. ;)
00:50:09Genre9mp3I think we all agree
00:50:22Genre9mp3Now as Davide said... we need a Dev! :)
00:50:24jhMikeSChanging the key map is all that's really needed
00:50:50 Join mirak [0] (
00:51:54 Join Febs_ [0] (
00:51:56jhMikeSas for the other stuff you're talking about, I"m confused
00:54:10Genre9mp3jhMikeS: What exactly?
00:54:33*jhMikeS pinched the pfc50606 asm i2c code to use on the x5 but has to figure out why it's not working...delays changes shouldn't change
00:54:53jhMikeSGenre9mp3: buttons assignments
00:55:42 Quit obo ("bye")
00:56:30 Quit ender` (" A psychiatrist is a fellow who asks you a lot of expensive questions your wife asks for nothing.")
00:56:44 Quit mirak (Remote closed the connection)
00:58:44Genre9mp3jhMikeS, File Browser: hold Rec => Recording Screen / tap Rec => queue file / Play+Rec => insert, WPS: hold Rec => Recording Screen / tap Rec => Show Current Playlist / Play+Rec => ???
00:58:46Davide-NYCThis brings up a good point: where is a wiki page (comprehensive) of button assignments?
00:58:58Genre9mp3jhMikeS: That was the idea
01:00:00 Join RogerBacon [0] (
01:02:15 Join holst [0] (
01:02:34holstmy h120 makes a "jitter sound" in the latest rockbox. is that a known issue?
01:02:52holstit was previously an issue with the orig. firmware from iriver, however they fixed it
01:02:53jhMikeShmmm...on x5 there'd be no menu access then since the rec button is menu and long rec = quickscreen from most places.
01:03:16Davide-NYCholst: try clearing your settings
01:03:32Davide-NYCif not let me know as I had this happen but it went away
01:03:57Davide-NYCjhMikeS: I'm going to go back to the office but instead of working I'm going to test your patch
01:04:09Paul_The_Nerdholst: Do you mean, via the remote?
01:04:12jhMikeSDavide-NYC: good use of time at the office imo
01:04:16Davide-NYCwhat exactly would you like me to do? Same as last time?
01:04:37holstPaul_The_Nerd: i will check if the sound disapears if im not using the remote
01:04:38jhMikeSCheck all the sample rates witht the plugin too and try recordin at them all. Don't have to test all formats
01:04:47jhMikeSJust use pcm wav
01:05:01 Quit Febs (Read error: 110 (Connection timed out))
01:05:11*jhMikeS will do an AIFF codec soon
01:05:14Davide-NYCwhat about channels?
01:05:20Genre9mp3jhMikeS: the idea was for the irivers (h100/h300) were the Rec button currently does nothing
01:05:35Davide-NYCNo man that doubles the work! Is it necessary?
01:05:37 Nick Febs_ is now known as Febs (
01:05:43holstyes it went away then
01:05:51holstthe sound is only there if im using the remote
01:05:54holstno thats no Yay
01:05:58Davide-NYCthat's a different issue
01:05:59holsti want my remote
01:06:08jhMikeSGenre9mp3: ahh...on x5 radio screen the long record could still be used to go right to recording.
01:06:20Davide-NYCis the LCD on your remote is updating it produces a f*ck*ng ticking noise
01:06:36holsta issue fixed on the orig. firmware
01:06:47Davide-NYCThis is unavoidable but can be reduced by turning on "reduce ticking" in the menu somewhere
01:06:50 Join NickDe [0] (
01:06:51jhMikeSDavide-NYC: really I'm concerned that all sample rates available work for rec/play
01:06:52holsti even reported it like 2 years ago to iriver ;)
01:07:02Davide-NYCthe way to fix this is to not update the display on the remote
01:07:24 Join TerrorByte [0] (i=4889bb7a@gateway/web/cgi-irc/
01:07:27Davide-NYCif you have peakmeters enabled (these update a lot) you'll here it during quiet passages
01:07:29holsthmm, not really a fix in my world, i guess i must deside: update or flac
01:07:29jhMikeSThe mono works and is really software only
01:07:48Davide-NYCjhMikeS: so no need to test channels
01:07:54Paul_The_Nerdholst: Just enable the "reduce ticking" option.
01:07:56jhMikeSno, those are fine
01:08:11Davide-NYCwill report back either here or in the forums.
01:08:16Paul_The_Nerdholst: The issue is a hardware flaw, so the only solutions are working around it, it can't actually be fixed entirely. The iRiver firmware update surely works around it as wlel.
01:08:31Davide-NYC^^ what paul said ^^^
01:08:37jhMikeSSee, them problem sometimes is that people "know" how the problem is fixed rather than magically being fixed by the manufacurer.
01:09:05*Davide-NYC still wishes the H1x0 had a RTC.
01:09:20jhMikeSThe assumption seems to be that the manu's solition is better.
01:09:35Davide-NYCIs there a way to list what patches have been applied?
01:09:51jhMikeSDavide-NYC: bummer...just tie your watch to it.
01:10:24Davide-NYCjhMikeS: I really get annoyed at not having my recorded convos not be named via datestamp
01:10:26jhMikeSDavide-NYC: no, why? Write them does
01:10:39jhMikeSWrite them down rather
01:10:51preglowman, usb is behaving erratic here right now
01:10:52*Davide-NYC could curse, but won;t
01:11:16 Quit MarcoPolo (Remote closed the connection)
01:11:29preglowevery time i insert usb now, rockbox spins up, spins down, THEN enters the usb screen, which spins the drive up again
01:11:30jhMikeSDavide-NYC: then say the say and time into the mic :)
01:11:33preglowwhat's up with this?
01:11:41*amiconn reappears
01:11:50Davide-NYCjhMikeS: that what I do *when I remember*
01:12:04 Join Rincewind [0] (
01:12:12Davide-NYCMan of the hour!
01:12:22amiconnpreglow: Are some of your questions from ~4 hours ago still open?
01:12:51jhMikeSamiconn: hmmm...I have one open too. Don't know if you can help. :)
01:12:58Davide-NYCRincewind there was just a raging discussion ( not really raging) about hard coding the Rec button functions. THis will affect your patch should it be implemented.
01:13:10*amiconn scrolls through backlog again
01:13:20*jhMikeS can repeast
01:13:32preglowamiconn: yup
01:14:07 Quit _Veseliq_ (Connection timed out)
01:14:12preglowamiconn: also, with 24 bit lpc routines, not only does my file play without skips, it now doesn't boost either
01:14:37 Join _Veseliq_ [0] (n=veseliq@
01:15:04linuxstbpreglow: Are you planning an ARM version?
01:15:08preglowlinuxstb: i am
01:15:09jhMikeSWas wondering if there's any other gotchas in pinching the better i2c driver code
01:15:12preglowPaul_The_Nerd: seems your powers aren't persistent
01:15:27Paul_The_Nerdpreglow: I was distracted, off ordering a replacement LCD remote for my H120
01:15:30Paul_The_NerdI seem to break them far too easily.
01:15:31*linuxstb gives Paul_The_Nerd some spinach
01:15:42Paul_The_NerdThey're down to $22 US though
01:15:43jhMikeSamiconn: It doesn't look like delays have to change. just the particular ports
01:16:48holstwhere is this "reduce ticking" option?
01:16:53 Quit spiorf (Remote closed the connection)
01:16:54amiconnjhMikeS: You're referring to [00:54:37] * jhMikeS pinched the pfc50606 asm i2c code to use on the x5 but has to figure out why it's not working...delays changes shouldn't change
01:17:35 Quit RogerBacon ()
01:17:52jhMikeSamiconn: yes
01:18:16jhMikeSmaybe I just made a dumb error but was wondering if there could be something else.
01:18:47TerrorByteme things Mike should go with latter thought.
01:19:03Mode"#rockbox +o preglow " by ChanServ (ChanServ@services.)
01:19:17 Join barrywardell [0] (n=barrywar@
01:19:52amiconnHmm, don't remember exactly. More than a month ago I had a look at adapting the pcf i2c driver for x5. Iirc there was some non-trivial problem
01:20:02jhMikeSI added the recalc for delays and all that
01:20:56*jhMikeS hates "non-trivial" in this context :0
01:21:22linuxstbamiconn: Do you still have the ATA test plugin you used when writing the SH/Coldfire ATA read/write functions?
01:21:53jhMikeSSuppose I should compare what i2c_generic does to what that does and adjust
01:21:53linuxstbI want to work on ARM verisons.
01:22:25amiconni2c_generic was done by Linus
01:22:36Davide-NYCholst: menu−−>General−−>Display−−>Remote-LCD−−>Reduce TIcking
01:23:38 Quit _flik_ ()
01:23:43amiconnThe idea itself would be good - if the performance wouldn't suck big time
01:23:58jhMikeSlol...yeah, but I'm need to know what the diff is in how the two handle things at the bit level
01:24:03amiconnIt's an inherent problem - either we do it generic, or we do it fast
01:24:09*jhMikeS likes fast
01:24:45jhMikeSIf it really helps so much it shouldn't really wait imo
01:25:01amiconnYes, at least for this driver which is called 100s of times per second
01:25:23*TerrorByte etyBrorreT
01:25:35TerrorByteEhm, anyways...
01:25:38TerrorByteJust wondering.
01:25:47TerrorByteThe full Doom .wad file.
01:25:50markunlinuxstb: I've just examined the firmware of the Gigabeat S
01:25:52TerrorByte'Tis freeware?
01:25:53amiconnjhMikeS: What are the pcf i2c port bits on x5?
01:26:00Davide-NYCI'm confused about this prompt: The next patch would create the file apps/enc_config.c, which already exists! Assume -R? [n]
01:26:05*amiconn could check the port pin page...
01:26:13Mode"#rockbox +b #rockbox!*@* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:26:22Mode"#rockbox -b #rockbox!*@* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:26:28Davide-NYCIs it asking me to assume whether or not I have reversed the patch or what?
01:26:38Mode"#rockbox +b #rockbox!*@* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:26:38markunlinuxstb: Looks like normal Windows CE, and I've found a tool to create a custom .bin file to load linux, maybe a good start for a rockbox bootloader.
01:26:40Davide-NYCtotally dumbfounded or simply dumb
01:26:40Mode"#rockbox -b #rockbox!*@* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:26:47jhMikeSSDA = 0x00010000, SCL = 0x00000400 I believe :)
01:26:58jhMikeSSDA = 0x00001000 rather
01:27:01amiconnYeah, but which ports?
01:27:12preglowsomeone hit me over the head with a stick until i can remember syntax
01:27:27Mode"#rockbox +b *!*4889bb7a@gateway/web/cgi-irc/* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:27:35Kick(#rockbox TerrorByte :preglow) by preglow!n=thomjoha@rockbox/developer/preglow
01:27:36amiconnNm, found it
01:27:41*Davide-NYC beats preglow with recently deceased fish about the head and neck
01:27:42jhMikeSSDA = GPIO1 and SCL = GPIO
01:28:00amiconnSo same situation as on H300, just different bits
01:28:20 Join webguest26 [0] (i=4670a5e6@gateway/web/cgi-irc/
01:28:21holstobviously, you had to have the remote inserted for the menu to be visible :)
01:28:31 Part webguest26
01:28:47jhMikeSI could have reversed something somewhere...I'll double check
01:28:53Mode"#rockbox -o preglow " by preglow (n=thomjoha@rockbox/developer/preglow)
01:29:08markunamiconn: what about keeping i2c generic and linking with inline functions, or wouldn't that work?
01:29:10linuxstbmarkun: Is info about the hardware available anywhere? Is there much in common with your gigabeat?
01:29:19 Quit TeaSeaLancs (Remote closed the connection)
01:29:30amiconnmarkun: You can't link inline functions, they have to be compiled in
01:29:45amiconnThat's how inlining works
01:29:52markunlinuxstb: don't know much about the hardware yet. The CPU is a ARM+DSP
01:31:03linuxstbDo you own an S, or are you just exploring the firmware?
01:31:14jhMikeSI'd think you could use the same driver and just #define the which ports and bits are used really.
01:31:34 Join webguest10 [0] (i=d8a58fda@gateway/web/cgi-irc/
01:31:42 Quit webguest10 (Client Quit)
01:32:14amiconnjhMikeS: That would only work as long as the general assignment is the same, e.g. SCL on GPIO and SDA on GPIO1 or vice versa
01:32:18markunlinuxstb: I don't own one
01:32:37amiconnIf both are on the same port, optimisation would lead to rather different code
01:32:46markunlinuxstb: but told someone how to hack his DAP :)
01:32:52*linuxstb gives up browsing's flash website
01:33:04 Join webguest10 [0] (i=d8a58fda@gateway/web/cgi-irc/
01:33:10jhMikeSThey're given to the assembly as named inputs anyway
01:33:17amiconnCompare lcd-h100-remote.c and lcd-remote-x5.c for an example of what I mean
01:33:27webguest10Anyone know where on the rockbox webpage I can find the different themes,wps...
01:33:38markunlinuxstb: here are 2 components of the gigabeat S
01:34:02amiconnjhMikeS: There are completely different optimisation options when both signals are on the same port
01:34:47Paul_The_Nerdwebguest10: The WpsGallery wiki page.
01:34:47markunlinuxstb: damn > 500MHz!
01:35:29preglowamiconn: got answer for my 21:35 question?
01:35:53amiconnThe latter would be faster for sure
01:36:26 Quit Kitt0s (Read error: 110 (Connection timed out))
01:36:36linuxstbmarkun: And a vector floating point co-processor...
01:36:43markunyes, sick
01:36:59linuxstbThat's just lazy...
01:37:10amiconnadd rx, (a0) does a read-modify-write, and add rx, ry; move.l ry, (a0) only adds & writes
01:37:47amiconnadd rx, (a0) : 3 cycles
01:37:55amiconnadd rx, ry : 1 cycle
01:38:08linuxstbThat's the kind of hardware more suited to Linux and an ffmpeg port than Rockbox...
01:38:08amiconnmove.l ry, (a0) : 1 cycle
01:38:15z0dejhMikeS you still about?
01:38:33jhMikeSz0de: yes
01:38:52z0dedid you see my question about alarm clocks and ubs/charging?
01:39:12jhMikeSz0de: no, sorry
01:39:54jhMikeSamiconn: of course like flipping bits simultaneously or whatever
01:39:55z0derepeat: I have the doc(x5) which has both usb and charger pluged in to it. can rockbox prioertise the charger?
01:40:14z0dei was told you were one of the people to ask.
01:40:15amiconnjhMikeS: Yes, with my beloved eor!
01:40:46amiconnMaybe it wouldn't be as helpful for i2c as it is for spi though
01:40:55jhMikeS:) ^ :( = )
01:41:22amiconni2c clearly defines when the data line must be flipped
01:41:53jhMikeSThat's something I have to learn myself, is the spec in the wiki?
01:42:03Mode"#rockbox +o preglow " by ChanServ (ChanServ@services.)
01:42:17Mode"#rockbox -b *!*4889bb7a@gateway/web/cgi-irc/* " by preglow (n=thomjoha@rockbox/developer/preglow)
01:42:20z0deof the x5?
01:42:28***Saving seen data "./dancer.seen"
01:42:46Mode"#rockbox -o preglow " by preglow (n=thomjoha@rockbox/developer/preglow)
01:43:37amiconnjhMikeS: I don't know about official i2c specs, but a lot of datasheets for chips using i2c communication have a protocol description
01:43:40preglowis there some interface for kicking people off the web client if that's needed?
01:43:49jhMikeSz0de: I don't think so...
01:44:27z0dek, it may actually be a iaudio thing as thats used for charging no rockbox logo in sight.
01:44:50jhMikeSz0de: If it's plugged into both, you get the usb connect too and charging through the adapter.
01:44:50z0dejust seems a bit daft to have a doc and then still have to unplug bits
01:44:52Davide-NYCOK I'm off (finally)
01:44:52 Quit webguest10 ("CGI:IRC (EOF)")
01:45:00 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
01:45:07z0deit charges via usb?
01:45:13preglowi ALWAYS get elusive bugs in the general lpc cases
01:45:17amiconnjhMikeS: Hmm, I can't see any pitfalls in pcf50606-x5.c now. Maybe I confused sth. Plugging the asm driver instead of i2c-generic should work
01:45:41jhMikeSz0de: that too if the usb port has the current capacity and it's not plugged into the AC adapter
01:46:07 Join jaebird [0] (n=jae@
01:46:27jhMikeSamiconn: Then I just made a dumb mistake. When I started it up, battery showed empty I don't think I has buttons.
01:46:31barrywardellis hcs still here?
01:46:43preglowwell, he's still in here
01:47:08hcsbarrywardell: I just got back
01:47:24hcs5 crashes
01:47:28barrywardellhey. just catching up on the irc logs
01:47:29preglowamiconn: do you have any idea why disk spuns up and down before the usb screen is displayed these days? it's damned annoying
01:48:03barrywardellhcs: the patch i made should be the same as how the h10 original firmware does frequency scaling
01:48:21hcshmm, guess there's something we need to consider for the ipod
01:48:36barrywardelli got occasional crashes on the h10 too
01:48:45barrywardellbut much much less frequent
01:48:45hcsbarrywardell: with which patch?
01:48:59barrywardellyes, after i posted my comment
01:49:11barrywardelli got the first crash
01:49:27barrywardellwithout the patch, i get crashes within a couple of minutes
01:49:34barrywardellwith the patch, it's more like hours
01:49:41 Quit jaebird (Remote closed the connection)
01:50:09hcssomething interesting I noticed was that I get no crashes when using dan_a'
01:50:18 Join jaebird [0] (n=jae@
01:50:20hcs*dan_a's cop patch
01:50:40 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
01:50:41hcsalthough it doesn't work exactly right, it seems to be helping something
01:51:46barrywardellyes, maybe we need to take the cop into account to avoid our crashes
01:52:14barrywardellhcs:Is that dan_a's patch + cpufreq4?
01:52:25hcsbarrywardell: no, just dan_a's patch by itself
01:52:38hcsran it for a whole weekend with no freeze
01:52:39 Join Landus [0] (
01:52:56barrywardellhmmm. interesting
01:53:07barrywardelli can't get dan_a's patch to work with my h10 :(
01:53:35hcsdoes cpufreq3 work on your h10?
01:53:48hcswith modifications to actually use scaling, that it
01:54:25*jhMikeS should check w/o the asm code
01:54:42*amiconn wonders why he didn't define the SCL and SDA bits as macros :/
01:54:48barrywardellyes, it seems to work quite well
01:54:57hcsoops, I missed a COP_CTL in 5
01:55:00barrywardelli think i might have still got a freeze, but i'm not sure
01:55:21barrywardellit works about the same as cpufreq4 i think
01:56:50amiconnjhMikeS: Hmm, did you take the different port setup into account?
01:57:15jhMikeSyes...changed it and defined all of it as macros :)
01:57:42amiconnThe H300 driver switches the ports to input for generating high level
01:58:10 Quit _Veseliq_ (Read error: 113 (No route to host))
01:58:25 Join lostnihilist [0] (
01:58:38amiconnThis is a bit more efficient because checks e.g. for clock stretching don't require extra direction switches
01:58:40barrywardellhcs: sometimes i wonder if scaling is meant to be done at all like rockbox does it
01:58:53barrywardellapparently the pp5020 has an automatic idle mode
01:59:00amiconnHmm, I could imagine one reason for the driver not working...
01:59:29hcsI've been meaning to get farther with my disassembly of the apple firmware...
01:59:31amiconn...if SCL and SDA have no pullups...
01:59:58barrywardellhcs: did you see the document i based my cpufreq4 patch off?
02:00:06jhMikeShmmm...will have to think* about that
02:00:12hcsbarrywardell: yes, it looked familiar
02:00:50barrywardellthe only thing i left out was the bit at the end
02:01:02barrywardellthe bit where frequency > 66MHz
02:02:10amiconnjhMikeS: Linus might know
02:02:15 Quit matsl (Read error: 110 (Connection timed out))
02:02:24pagefaultwork on pp cpu scaling?
02:02:42barrywardellpagefault: yes
02:02:47pagefaultooh nice
02:02:57jhMikeSamiconn: for sure
02:03:08hcspagefault: how's SPC playback coming?
02:03:09pagefaultalthough with your disk shut off patch I get a nice battery boost
02:03:41z0deis it possible to make the x5 with rockbox show an image in the background depeneding on the song?
02:03:48pagefaulthcs almost ready for release
02:03:54barrywardellpagefault: yes, but i've heard of some problems when disk shut off enabled. haven't been able to reproduce them myself yet
02:04:01Paul_The_Nerdz0de: No, it's not.
02:04:02preglow24 bit flac playback works
02:04:05z0dee.g all my music has folder.jpg in the album.
02:04:08preglowwell below boost
02:04:11pagefaultbarrywardell, sometimes it does lock up the player it can't turn the disk back on
02:04:24pagefaulti've experienced it if I leave it idle
02:04:55jhMikeSiirc correctly they do
02:05:32barrywardelli'm not sure what the source of that problem is. it seems like something that should affect all targets with disk power off enabled.
02:05:53pagefaulthmm only h10 affected
02:06:16amiconnbarrywardell: I don't think so. It can be hardware dependent.
02:06:26jhMikeSI don't know, guess I'll ask later or if you see him. Haven't been around when he's here
02:06:35pagefaultperhaps the disk needs time to turn on back again
02:06:37 Quit dpro (Read error: 110 (Connection timed out))
02:07:29pagefaultthe disk spins up then it just locks up
02:07:40barrywardellamiconn: true, but the poweroff is only really two lines of code. maybe we're missing a step?
02:07:40pagefaultlike it was trying to read then crashed before the disk was ready
02:08:09hcsbarrywardell: I'll be testing again with cpufreq5, with that one line I'd missed removed, but I'm not expecting much of anything
02:08:24barrywardelli wonder if we should force a delay after re-enabling disk power???
02:08:36pagefaultit might be HD model dependant
02:08:43amiconnbarrywardell: Example: The archos player comes in 3 hardware versions. Version #1 has no disk power control at all. Version #2 _can_ power off the disk, but will crash when powering on again. Version #3 has no such problems
02:08:58pagefaultthe default iriver firmware gives it like 5 seconds to spin up
02:09:16amiconnVersion #2 is the reason why official rockbox builds for the player have no disk poweroff option. My local version has that option
02:09:41 Join aliask [0] (
02:09:43barrywardellhcs: i've pretty much followed what we're told the iriver firmware does word for word. i wonder if it's something else causing problems
02:10:10pagefaulti'm going to experiment with delays
02:10:11hcsbarrywardell: well, I can't stand to run it with the channels flipping all the time
02:10:21amiconnIf you have a closer look at the hardware of an arbitrary rockbox targets, you will find a bunch of hardware design error/drawbacks/whatever
02:10:35barrywardellamiconn: i see. so maybe disk poweroff isn't meant to be be done on the H10
02:10:37hcsand now, to dinner
02:10:59barrywardellit's an option that's defaulted to off anyway is suppose
02:11:20amiconnIt _should_ default to enabled thoguh
02:11:21barrywardellpagefault: great. let me know if you have any success.
02:11:47pagefaultit's a shot in the darek
02:11:52pagefaultI mean dark
02:11:59barrywardellamiconn: yes, in an ideal world :)
02:12:28pagefaultthese are ATA drives in the h10 right?
02:12:44barrywardellhcs: there are people on misticriver complaining of channel flipping even without frequency scaling
02:12:49barrywardellpagefault: yes
02:13:25pagefaultoh well I am off to play with code now
02:13:27pagefaultI have an idea
02:14:42amiconnbarrywardell: channel flipping?
02:14:47pagefaultit could be that when the disk is powered back on it is still in standby mode as well
02:14:55pagefaultI will look into it
02:16:04barrywardellamiconn: flipping of right-left in audio
02:16:22barrywardellpagefault: yes, quite possible
02:16:26amiconnWhen the disk is powered back on, it will _always be in standby_. The driver will activate it
02:16:44amiconnThis has to be done on every ata target
02:16:44pagefaultso does the current code take it out of standby mode
02:16:46 Quit Rincewind ("Bye")
02:16:54amiconnof course
02:17:17barrywardellthe H10 hdd seems to do some things all by itself.
02:17:38pagefaultdepends what the hell is on the firmware on the drive too yeah
02:18:03amiconnThe hdd firmware implements the ata protocol, which is standardized
02:18:27pagefaultI see
02:18:51amiconnIf the ata driver doesn't work correctly on a certain target, there must be a mistake in the very low level part for that target
02:19:11amiconnSome port setup missing/incorrect, or some out-of-specs timing
02:19:17linuxstbbarrywardell: Do you have a patch to fix the debug_menu.c problem on OS X?
02:19:19 Join _Veseliq_ [0] (n=veseliq@
02:19:45pagefaultI am just curious thats all, I don't really know much about this stuff
02:20:09barrywardelllinuxstb: I do, but not a very good one. it just disables pretty much everything in debug_menu.c
02:20:59pagefaultit would be cool if the sleep timer could be saved though
02:21:03linuxstbDo you also get size_t warnings?
02:21:10barrywardelli haven't had a chance to isolate the exact cause of the problem
02:21:16pagefaultlike you don't push any buttons or stuff for that amount of time and it shuts off
02:21:18amiconnHmm, the X5 pcf i2c should have pullups...
02:21:23pagefaultit's kind of annoying to set it each time
02:21:25barrywardelllinuxstb: with or without the patch?
02:22:18barrywardelli'll do a test compile now...
02:22:20 Join jborn_ [0] (
02:23:16barrywardelllinuxstb: nope, this is what i get:
02:24:05 Quit jaebird ("Ex-Chat")
02:24:09linuxstbWhich native gcc are you using? I seem to have 3.3 here.
02:24:28barrywardellpowerpc-apple-darwin8-gcc-4.0.1 (GCC) 4.0.1 (Apple Computer, Inc. build 5341)
02:24:48barrywardellso it's probably a probably a problem with apple's gcc
02:25:22jhMikeSamiconn: I think so...but _I_ could be confused. Asked him a bunch of q's way back in July about stuff.
02:25:34barrywardellit's a very strange error
02:25:55linuxstbbarrywardell: Yes, seems I need to upgrade. I get this error now: "/usr/bin/libtool: for architecture: cputype (16777234) cpusubtype (0) file: -lSystem is not an object file (not allowed in a library)"
02:26:48barrywardelllinuxstb: i think i'm running a recent XCode, which installs gcc with it
02:26:53linuxstb(linking vorbis.codec)
02:27:12preglowanyone on iriver that can test 24 bit flac playback?
02:27:44pagefaultdisk isn't coming back from sleep
02:27:58*amiconn has no idea how to get hold of 24bit flac files
02:28:22preglowi've only got a couple, and they sound nice
02:28:31linuxstbThere are some at, but their search was broken (at least it was an hour ago)
02:28:51pagefaultthis is weird
02:29:00pagefaultit literally shuts off the disk now and it doesn't turn back on
02:30:19*linuxstb hugs google:
02:30:39pagefaultbut connecting USB turns it back on fine
02:30:41amiconnHmm, lloks like we aren't powering off the disk on ipods... (?)
02:30:47pagefaultbut that just kicks me back to iriver's firmware
02:30:54pagefaultso it looks like it is possible
02:31:01pagefaultbut we just don't do it right
02:31:40amiconnpagefault: Just for clarification, are you talking about spindown or true poweroff?
02:31:41jhMikeSamiconn: are you talking about enable/disable being used for lo/hi instead of GPIO_OUT? That's what you mean?
02:31:42linuxstbpreglow: Are your 24-bit files 44.1KHz?
02:31:47pagefaulttrue poweroff
02:31:54pagefaulton the h10
02:32:04amiconnjhMikeS: Yes.
02:32:06pagefaulttrying to figure out why the disk never wakes up
02:32:27amiconnpagefault: Then you're doing something on the h10 we don't do on the ipods
02:32:49*amiconn wonders whether the hardware is the same or at least similar
02:32:51pagefaultthe hd wasn't turning back on then I did the USB trick and it spun up
02:32:52barrywardellamiconn: i found a gpio port which seems to cut power to the H10 hdd
02:32:55jhMikeSamiconn: the generic explicitly sets GPIO_OUT. perhaps should adapt it to work that way.
02:33:32amiconnjhMikeS: I know that the generic driver does that... but if you do, you'll lose quite some optimisation potential
02:33:52pagefaultamiconn, i'm just using the cvs code at this point
02:33:55preglowlinuxstb: yeah
02:34:01amiconnI tend to believe that you just overlooked something...
02:34:10jhMikeSamiconn: well...the non asm seemed to behave exactly the same way as the asm
02:34:13 Nick jborn_ is now known as JoeyBorn (
02:34:31*linuxstb will have a 24-bit/96KHz test track in about 30 minutes time...
02:34:39pagefaulton the h10 you get the old iriver builtin firmware to come back when you plug in usb
02:34:53pagefaultand that spins up the disk after rockbox has turned it off
02:35:44 Join jaebird [0] (n=jae@
02:35:52*barrywardell finds it very easy to hear what the h10 disk is doing!
02:36:04*amiconn should go to sleep now, otherwise he'd try to plug the asm code into the x5 pcf driver himself :/
02:36:06pagefaultyeah it's a high pitched noise
02:36:30jhMikeSamiconn: I see, then you have to start switching input<==>outputs explicitly.
02:37:21jhMikeSamiconn: nighty night
02:37:45jhMikeSamiconn: want a pastebin?
02:38:09amiconnjhMikeS: Correct, and you'll very probably run out of registers for hold addresses & masks in the loops...
02:41:40jhMikeSI think I spotted it
02:43:30barrywardelllinuxstb: found the problem with the mac os x sim
02:43:35barrywardell __attribute__ ((section (".icode")));
02:43:41barrywardellon line 393
02:43:54barrywardellremoving it makes the sim compile fine
02:44:21jhMikeSheh...working now
02:46:32barrywardellsurely that whole function shouldn't be compiled for the sim?
02:50:12jhMikeSogg boost goes from 78% -> 35% harrrmphh
02:51:12 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
02:51:20jhMikeS0% boost on mp3 using the EQ
02:51:34jhMikeSfrom 18%
02:51:43preglowamiconn: did you check if tvl and wolfson click as well?
02:52:09preglowjhMikeS: doing what?
02:52:28jhMikeSAdapting the pcf50606 driver that's used for iRiver to the x5
02:52:37jhMikeSThe i2c part
02:53:13jhMikeSThis is using the non-asm version. The ASM has something wrote but it shows how nescessary this is.
02:53:26preglowand you got that much boost improvement???
02:53:27preglowholy shit
02:53:40jhMikeSIt scrolls with no boost like it's boosted almost
02:53:52preglowwhat's the pcf used for all the time?
02:54:07jhMikeSEverything...RTC, buttons, battery,
02:54:32jhMikeSIt communicated with 100s of times a second
02:54:34 Join JdGordon [0] (
02:54:48jhMikeSNo boost using the EQ is cool
02:55:05pagefaultEQ makes me skip
02:55:06preglowhell yes
02:55:09pagefaultI wish I could use it
02:55:21preglowok, i can't be bothered to tinker around with this anymore, so i'm just commiting it
02:55:56jhMikeScommiting what? the 24bit thing?
02:56:25jhMikeSIt's like a different player now...this should have been done a long time ago.
02:57:02jhMikeSNow I get 54.5 fps on full screen updates too
02:57:15jhMikeSI'm gonna see what video plays at...
02:58:16 Join Davide-NYC [0] (
02:58:41Davide-NYCI did some tests
02:59:09Davide-NYCThe recording end seems to work, but I see you've disabled some of the 11kHz availablilty
02:59:10jhMikeSuh oh... :)
02:59:22jhMikeSDavide-NYC: from where?
02:59:48jhMikeSYou can't use 11kHz for MP3 because that's only for MPEG2.5
03:00:10jhMikeSIt should never have been there
03:00:11Davide-NYCOK. I will never object to removing 11kHz as an option
03:00:27jhMikeSDoes it work on WAV and for playback though?
03:01:44*jhMikeS will be able to fiddle with it himself by the 30th
03:02:04Davide-NYCThe WAVs playback perfect on the PC w foobar2k
03:02:19jhMikeSSo you got em in at all samplerates
03:02:37Davide-NYChold I'm listening back now
03:04:10Davide-NYCMic/Wav/Mono at all three SR no prob
03:04:11 Quit midkay ("Leaving")
03:04:31jhMikeScool...the plugin plays back ok?
03:05:09Davide-NYCYes AFAICT
03:05:21Davide-NYCnot even glitchy on changes
03:06:54Davide-NYCActually it's a bit glitchy in that I can hear some "shearing" between lower samplerates
03:07:03Davide-NYCbut only when switching
03:07:38 Join midkay [0] (n=midkay@rockbox/developer/midkay)
03:07:39jhMikeSI don't think it's a concern to be able to switch sample rates live without any glitch and probably unavoidable in most cases.
03:07:52Davide-NYCThat's what I call it.
03:08:16Davide-NYCI'm imagining some infinite slope sections of the waveform
03:08:23Davide-NYClike a vertical line
03:08:37Davide-NYCbut of course that's just my imagination
03:09:05Davide-NYCjhMikeS: check out my menu idea:
03:09:36Davide-NYCIt's all about IDCs... Input dependant Configurations
03:09:59Davide-NYCand a little bit about nested sub-menus.
03:10:19*Davide-NYC guesses that nested sub-menu is redundant
03:11:00jhMikeSnature calls...then will look...
03:11:10 Quit funky ("leaving")
03:12:41midkayDavide-NYC: nested submenus are inconsistent altogether with rockbox's current menu system.. and if it's just two entries it's pretty pointless to have a submenu at all.
03:12:56Davide-NYCMic/MP3/160kbps/Mono and Stereo at both samplerates sounds good to me as well.
03:12:57midkayi mean "long-hold-button-popup" submenus.
03:13:38Davide-NYCmidkay: I know, but where else are the menus dependant on another selection?
03:14:03Davide-NYCIf another scenario like this exists I'd like ot take a look at how it has been implemented.
03:14:07Davide-NYCLemme 'splain
03:14:09midkayah, i see what you mean.. submenus are fine, but not ones that expand if you hold a button. that's just inconsistent.
03:14:29Davide-NYCIf you chose MP3 your available bItrate and samplerate options change
03:14:57Davide-NYCif you choose a certain MP3 bitrate your samplerate options change
03:15:05midkayi see.
03:15:33Davide-NYCall of these things depend on one another. Should a similar situation exist elsewhare that better conforms with the RB menu paradigm I'd be interested in studying it
03:15:52*Davide-NYC must learn to type eventually
03:16:08linuxstbDavide-NYC: Why not just normal submenus? i.e. one for MP3 settings, one for WAV, one for Wavpack.
03:16:32midkayDavide-NYC: so you're arguing in favor of inconsistent menus, or just submenus? i've no problem with the latter..
03:16:43Davide-NYCJust sub menus.
03:16:59Davide-NYCread the forum post I think it's pretty clear
03:17:09Davide-NYCif not let me know and I'll make changes
03:17:21midkayexcept the part about, oh, inconsistent hold-button-to-expand submenus :)
03:17:28midkaywhich is what i wanted to clarify, everything else is clear.
03:17:38Davide-NYCI was thinking of it like htis
03:17:59 Quit |AhIoRoS| (Remote closed the connection)
03:18:03jhMikeSDavide-NYC: Was debating putting channels in the format submenus at the beginning and am still thinking about it.
03:18:07Davide-NYCshor click selects the encoder type and auto-inherits the settings "inside"
03:18:10jhMikeSNice glowing text BTW
03:18:48Davide-NYClong click opens the menu for modification of said inherited settings
03:18:56hcsbarrywardell: back from dinner, freezes as expected, I'm going to play around with the COP patch
03:18:57midkayDavide-NYC: but that's weird and kinda un-intuitive, and certainly inconsistent.. why not just MP3 Settings, WAV Settings, Wavpack Settings, and then Encoder Selection where you choose MP3/WAV/Wavpack.
03:19:02jhMikeSSome like it some don't but I'm really concerned about the need of those who have recording as a priority with this.
03:19:21Davide-NYCGuys let me explain the last paret of the post
03:19:30Davide-NYCIt's all about inheritance
03:19:48midkayand Encoder Selection would *inherit* [codec] Settings that you already set via standard submenus.
03:20:33Davide-NYCWell it's sort of "upward inherit"
03:20:56midkaywhich is a style of inheritance that i don't use in my suggested method of encoder/setting selection.
03:21:01Davide-NYCthe last bit in the post would be that the source you select inherits upward the settings you gave it last
03:21:10barrywardellhcs: did you experience any freezes with v3?
03:21:24midkayinstead i go for interface consistency and intuitiveness.
03:21:31Davide-NYCmidkay, how would you tackle the interdependance problem?
03:21:38hcsbarrywardell: once, I think, in the several weeks I used it
03:21:41*jhMikeS 's head is spinning at the thought...
03:21:50Davide-NYCI think it's awesome
03:22:07midkayDavide-NYC: didn't i explain? Encoder Selection chooses MP3, WAV or Wavpack. the selected encoder will then use the MP3, Wav or Wavpack settings you've specified on a per-codec basis.
03:22:20jhMikeSYou want sources to have encoder preferences
03:22:42Davide-NYCThat's the way it works now (with jhMikeS patch) and the way it would work under my system
03:22:48Davide-NYCthe only diff is the presentation
03:22:49midkaybitrate/frequency based on codec based on source? insane..
03:23:10Davide-NYCOK, take the source bit out
03:23:19jhMikeSsource selects preferred way to record it...makes sense to me really
03:23:25midkayDavide-NYC: exactly - it's a kinda-neat-but-inconsistent-and-requires-a-manual-to-figure-out presentation versus a standard-menu-system-we-use-all-over presentation.
03:23:45Davide-NYCBut midkay (in my best scotty voice)
03:24:00Davide-NYCMP3 dictates certain freq at certain bitrates
03:24:12jhMikeSand channels at certain bitrates :)
03:24:19Davide-NYCso it's based on encoder regardless
03:24:26midkayDavide-NYC: so use a "source preference" menu too where you have three entries - Mic, Digital, Line In and you set either MP3, WAV or Wavpack in each of those. problem solved.
03:25:01linuxstbThat could just be done with .cfg files... It seems too complex otherwise.
03:25:10Davide-NYCmidkay, have you patched with jhMikeS big recording patch?
03:25:11 Part pixelma
03:25:25midkayit should be all configurable in a standard menu system (or not at all, it's pretty complex) or it'll be weird to nearly everybody... i can nearly guarantee its rejection if you want to get funky with menu systems.
03:25:29midkayDavide-NYC: no, i don't record.
03:25:38jhMikeSIt's quite doable. Mr. Brain* already computed it.
03:25:44 Join AdamKili [0] (
03:26:06Davide-NYCnot be rude, but I'm pretty sure there is a piece of this domain you're not understanding
03:26:08jhMikeS* Not nescessarily a real brain
03:26:33Davide-NYCPlease patch and explore
03:26:41jhMikeS* A cow brain wired to an interface :P
03:26:46Davide-NYCand then generate a rebuttal post on the firum
03:26:58Davide-NYCjhMikeS: you get what I'm talking about?
03:27:02midkayDavide-NYC: i can't, no iRiver. what am i missing?
03:27:26Davide-NYCan enthusiasm and understanding of the recording dsomain
03:27:42jhMikeSYes. It's just a matter of shuffling settings around and once the recording code and samplerate code is up that can be shuffled around np.
03:27:43Davide-NYCand an dap that records
03:28:03midkayi don't tend to display enthusiasm towards weird new menuing systems for only very certain areas of the menu where everything else uses a single other method...
03:28:17Davide-NYCyou're missing the forest for the trees
03:28:17jhMikeSThe encoder menus are not locked to a certain context and can be displayed anywhere
03:28:52Davide-NYCjhMikeS: what do think about persisten input dependant configs?
03:28:52midkayDavide-NYC: stop with the "you can't record so you don't know what i mean" stuff. i understand perfectly. it's just too weird versus the rest of rockbox.
03:29:24Davide-NYCI'm not on about that. I'm saying that the recording enthusiast has different requirements.
03:29:25AdamKilihmm.....Creative Zen Vision: M, or rockbox'd 80 GB iPod 5.5 (when supported)?
03:29:31hcsbarrywardell: so you say dan_a's patch doesn't work on the h10?
03:29:42jhMikeSDavide-NYC: I think if the ones who do a lot of recording would find it useful I like it too :)
03:29:47midkayDavide-NYC: i think requirements have nothing to do with using a weird menuing setup.
03:30:01midkayit might require about two more clicks to achieve the same thing.
03:30:03linuxstbAdamKili: You would only get one answer here - any Rockbox'ed device.
03:30:04midkaywhat's the big deal?
03:30:31barrywardellhcs: yes, i don't get two main threads, just the one on the main cpu
03:30:37*jhMikeS knows some of these points were controversial so he preserved the current Encoder Settings/Frequency/Channels setup.
03:31:04Davide-NYCIf you fully understand the domain (encoder restrictions) and can generate an interependant menu system that adheres to the rockbox menu paradigm please put up a rebuttal post in the forum.
03:31:09 Join midgey34 [0] (
03:31:12AdamKililol ya. just can't decide
03:31:14Davide-NYCI don;t see it
03:31:23midkayi don't think mutilating and destroying rockbox's nice consistency, simplicity and intuitiveness is worth saving a couple clicks.. especially when you might need to totally stop to read the manual - that certainly negates the few clicks you save.
03:31:40hcsbarrywardell: could you try it with a kernel running on the COP but just starting all threads on the CPU? I'm going to try out a patch that does that here
03:31:47linuxstbbarrywardell: I got the same the first time I tried it on my 5g - just a codec thread. But I tried it a few days laters, and got a main thread, so it's a mystery to me... I tried lots of earlier versions of the COP patch on my Photo, and never got a COP thread running.
03:31:48midkaysure, i iwll.
03:32:09jhMikeSI don't think source dependent settings should remove the current global setup though.
03:32:33Davide-NYCmidkay your point is valid, and I think you should put up a forum post using similar "glow" formatting as I did so that the two designs can be properly discussed
03:32:39jhMikeSThey would use the global settings until explicitly set somehow.
03:32:41midkayi am.
03:32:48barrywardellhcs: can you provide me with a patch and i will try it?
03:33:16Davide-NYCjhMikeS: so yet another menu?
03:33:20linuxstbjhMikeS: The standard Rockbox way to handle things like that is allowing the user to load .cfg files with the settings - so users can set them up for different recording environments.
03:33:21jhMikeSThat way no behavioral change takes place
03:33:29barrywardelllinuxstb: yes, i noticed that. seems there is still some work to be done on the patch
03:33:45Paul_The_NerdDavide-NYC: I don't understand, what qualifies as an "independent" setting anyway?
03:34:06barrywardelli've been trying to get jtag working with my h10. it could make debugging this stuff a lot easier!
03:34:20jhMikeSUnder source options: "Use Global Settings" or the like would be the default.
03:34:41barrywardelli'm still looking for good (free) software for using jtag with arm
03:34:44Davide-NYCPaul_The_Nerd: Codec independant = Source, Channels, File split, etc..
03:34:59Davide-NYCdependant = encoder, bitrate, samplerate
03:35:03Paul_The_NerdDavide-NYC: So, why are specialized menus necessary at all?
03:35:18jhMikeSChannels and Frequency are currently the only settings whose selection depends on file format
03:35:22Paul_The_NerdOnce you pick an encoder, why not just have the available bitrate and sample rates show different lists depending on the encoder?
03:35:26linuxstbbarrywardell: Try finding davidc in #ipodlinux or #ipodlinux-dev - he's succeeded in using JTAG with a PP5020 ipod.
03:35:50Paul_The_NerdHave sane defaults so that if someone changes the encoder without manually changing the others, it resets to something semi-happy.
03:35:51linuxstbbarrywardell: He's got a wiki page on about it as well.
03:35:59 Part AdamKili ("Ex-Chat")
03:36:05Paul_The_NerdBut I don't understand any benefit of a custom menu structure rather than simply a standard settings menu
03:36:22jhMikeSIf you have WAV selected at 11kHz and the you select MP3 you'll find your Frequency automatically set to 22kHz for example
03:36:25Davide-NYCBecause visual heirachy should relate to logical heirarchy?
03:36:37hcsbarrywardell: , its just the kernel_on_cop_4.diff without the patch to run the codec thread on the COP
03:36:43Davide-NYCHey, if I'm the only guy for this it's a dead point
03:36:58Davide-NYCbut I'd like it discussed at some length first
03:37:24barrywardelllinuxstb: thanks. i'll ask him about it
03:37:30Davide-NYCBut, if I can generate a config file now that sets up my rec settings based on source, I'll shut up and just make one
03:37:43*jhMikeS should probably commit the i2c driver change w/o the asm enabled since the improvement is so huge.
03:37:48Davide-NYCand forget the whole ugly incident. ;-)
03:37:54linuxstbIs 11.025KHz MP3 the only problem? If so, then we should try and fix the MP3 encoder to handle it.
03:38:14Paul_The_NerdIt seems to me that it'd be logical to save independently for each source, what you've got the settings at, but I don't think custom menuing is at all necessary.
03:38:23jhMikeSlinuxstb: no...if you have 88.2 WAV selected it's the same sort of thing.
03:38:29barrywardellhcs: i'll try that now...
03:38:31 Join AdamKili [0] (
03:38:36jhMikeSNo MPEG L3 goes to 88.2
03:38:41Davide-NYCPaul_The_Nerd: please say so in the forum
03:39:12midkaythere, i replied...
03:39:21Davide-NYClinuxstb: 11kbps MP3 sound so aweful I say leave it alone.
03:39:22barrywardellhcs: i just remembered. enabling cpu freq adjust with the cop patch caused all sorts of weirdness
03:39:25Davide-NYCmidkay thanks
03:39:28jhMikeSlinuxstb: would be nice to have MPEG 2.5 available too
03:39:46jhMikeSDavide-NYC: 11kHz!
03:39:46hcsbarrywardell: well, it doesn't seem to boot anyway
03:40:05AdamKiliis there a galery somewhere that shows screenshots of every menu and submenu of rockbox on a 5G ipod? so that i can get a good idea of what it would be like and what it's exact features are?
03:40:07 Join SmilinBob [0] (
03:40:44jhMikeSCompressed formats have their own set of acceptable inputs. The ones that don't care would be WAV, WavPack, and say AIFF.
03:40:45scorcheAdamKili: there is the manual
03:40:58AdamKilioh i'll check it out
03:41:09scorcheif you want features, there is the WhyRockbox page in the wiki also
03:42:04jhMikeSOk...recording presets! There!
03:42:10Paul_The_Nerdmidkay: I posted my thoughts as well, I think it's a simplified extrapolation from yours.
03:42:27AdamKilican u give me linky to whyrockbox page in the wiki?
03:42:31***Saving seen data "./dancer.seen"
03:42:34AdamKilii can't seem to find it
03:42:43linuxstbIt's linked from the top of
03:42:46jhMikeSI gotta get that i2c asm working...and fast! :)
03:43:14AdamKilinvm found it
03:43:20Davide-NYCmidkay: Your method does not allow for chosing an encoder
03:43:23midkayPaul_The_Nerd: that's entirely what i'd like to see. mine was based upon this apparently huge demand for settings based upon settings based upon settings.
03:43:35midkayDavide-NYC: add a Source menu then. no big deal.
03:43:37Davide-NYCPaul_The_Nerd: I guess you way is most Rockbox consistent
03:43:44scorcheAdamKili: and you can always go into the wiki and type the name of the page you want to go to in the box at the top
03:44:48AdamKiliit didn't find it that way. that's y i was confused. maybe i typo
03:44:51Davide-NYCI still think visual heirarchy should relate to logical heirarchy but if I'm alone (and I'm not convinced I am until the recording comunity chime in) I'll drop it.
03:44:51AdamKilid it
03:44:56midgey34barrywardell: i think i fixed the mac sim issue
03:45:05Paul_The_Nerdmidkay: Mine is somewhat codependent, but I think in a more minimalistic way. :)
03:45:29Paul_The_Nerdmidkay: I really like the idea of populating Bitrate and Frequency based upon what Format is set at, but don't know how possible that is with current menuing code.
03:45:30midkayDavide-NYC: a sincere good-luck with getting any kind of expanding menu system committed if that's the direction you guys take. :)
03:45:34barrywardellmidgey34: check my latest cvs commit ;)
03:45:56midkayPaul_The_Nerd: ah, i see.. that's a very good way to look at it.. mine was kinda a direct translation of the expanding menus into a submenu system.
03:46:06Davide-NYCwhy, where else would there be benefit from expanding sub-menus?
03:46:53midkayDavide-NYC: nowhere else. there'd only be a benefit from not having them - no missed recordings due to a required manual read by most people, no consistency or intuitivity complaints.
03:47:03Paul_The_Nerdmidkay: Have you given any thought to adapting the clock sliding highlight bar an option for the main menu code, like that one person was pestering about? (It was you that did it, right?)
03:47:10barrywardellmidgey34: looks like we did things differently though. you fixed the problem, I just avoided it!
03:47:20Davide-NYCmidkay please don't type sincere when you are not/
03:47:24Davide-NYCthat's just whack
03:47:27Davide-NYCand unnecessary
03:47:49midkayDavide-NYC: no, i was sincere. you're gonna need a lot of good luck to get something like that in. :)
03:47:50Davide-NYCI'm trying to figure something specific out for the benefit of people that record and for the benefit of RB
03:47:53midgey34barrywardell: well, i dont know if my problem causes problems on other OSs
03:48:14midkayPaul_The_Nerd: yes - as soon as I get back into rockbox coding i was planning on taking a crack at some animations like that.
03:48:31jhMikeSmidkay: expanding menu system? you mean explicit tree with [+]|[-]?
03:48:44barrywardellmidgey34: it might be worth committing your suggested changes too, but i'm don't know enough to be sure
03:48:44 Quit AdamKili ("Ex-Chat")
03:48:57midkayjhMikeS: i mean short/long presses like Davide-NYC illustrated.
03:49:09*jhMikeS will be banished from the land
03:49:15midkayDavide-NYC: is it a benefit if someone misses something crucial because they can't figure out why these menus aren't opening?
03:49:27midkayor how they work at all even if they get them to open.
03:49:34Paul_The_NerdI think that menu navigation should *never* require long vs short presses.
03:50:01Paul_The_NerdMenu navigation is a pretty basic concept, and should be kept as absolutely uncomplicated as possible
03:50:07linuxstbHow about moving all the recording settings into the codec settings? i.e. Recording Format (mp3/wav/wavpack), and then "mp3 settings", "wav settings", "wavpack settings" which include source, samplerate, channels etc.
03:50:30midkayconsistency and intuitivity is important. all rockbox menus can be used the same way - in, adjust, out. if you go and set up one menu to use long presses and expanding submenus, that consistency is lost.
03:50:44Paul_The_Nerdlinuxstb: I think you could get away with just setting the currently selected codec's settings, if there were some way to query the codec for what valid options were for each.
03:50:44jhMikeSI think showing a setting at the right and changing it in place like many OF and appliances do would be nice though.
03:50:57Paul_The_Nerdlinuxstb: I think that would then be more easily extendable without having to explicitly expand the menu for new formats.
03:50:58barrywardellhcs: with that patch i just get all threads running on the cpu, and just the one main thread
03:51:02midkayjhMikeS: that's an entire firmware change, though.
03:51:28midkaysomeone coded it actually.. i've seen it on the forums. it's nice, but certainly not for a single menu.. that's a very WTF situation to come across when everything else behaves another way.
03:51:40linuxstbI just think it's confusing that changing codec (or something else) will magically change another setting.
03:52:13hcsbarrywardell: I messed up, it won't even start up on my ipod
03:52:31linuxstb(e.g. if you have 11.025KHz or 88.2KHz selected, but then change the codec to MP3)
03:52:50*Paul_The_Nerd shrugs
03:53:04Paul_The_NerdIt seems to me it'd be fair that when you select MP3, it defaults to sane MP3 settings that you can then change.
03:53:10jhMikeSlinuxstb: yes...a bit. Like I said, that can be shuffled around rather easily though.
03:53:23Paul_The_NerdBut I'm not a big recorder, I was just thinking for menu neatness, not having duplicated options for every single codec
03:53:55*jhMikeS points out why he left it alone as much as possible :D
03:54:12Davide-NYCI totally see now
03:54:54*barrywardell needs sleep
03:54:59 Quit barrywardell ()
03:55:53*jhMikeS isn't sure what MP3 settings are "insane". It just defaults to the nearest matching setting and the last encoder specific settings used.
03:56:03 Quit _Veseliq_ (Read error: 110 (Connection timed out))
03:56:18Davide-NYCMP3/8bps/11kHz/Stereo 'was' insane
03:56:26Davide-NYCrobot noisews
03:56:41jhMikeS8bps is pretty insane! Not sure where you got the encoder :)
03:57:08Davide-NYCwait, (lol) what the lowest value for the first version of your patch?
03:57:09jhMikeSNot to mention have 11kHz with MP3
03:57:52 Join _Veseliq_ [0] (n=veseliq@
03:57:59jhMikeSStill is but I may just restrict those low bitrates to mono since 16kbps stereo = 2*8kbps mono
03:58:22Davide-NYCI have a file of myself speaking where I have to presume it says "MP3, 8 kilobits per second, 11 kiloHertz, Stereo"
03:58:43Davide-NYCNo way of knowing otherwise, unless you're a droid
03:58:50jhMikeSThe lowest samplerate available for the MP3 is 22050
03:59:00Davide-NYCI say leave it
03:59:19jhMikeSI'm versed in over 6,000,000 forms of communication
03:59:34Davide-NYChelp me jhMikeS, y9ou're ou only hope!
03:59:42*Davide-NYC must learn to type
04:01:08Davide-NYClinuxstb: you earlier mentioned that you found it 'wierd' that your settings could change based on other choices. If you have a design idea to alleviate that please post it in the forum.
04:01:09jhMikeSI think having a complete useless and unintelligible setting is a bit "insane"
04:01:20Davide-NYCThat's what I like to hear
04:01:56jhMikeS8kbps is quite intelligible for mono though so I don't have problem with that combo
04:02:23jhMikeSIf it's useful for voice only it's good
04:02:47Davide-NYCOh dear, you didn;t restrict it though
04:02:57jhMikeSnothings restricted yet
04:03:01Davide-NYC8kbps/22k/stereo is beeps and boops only
04:03:15Davide-NYCyou're telling me mono is OK?
04:03:34jhMikeSYeah...for voice.
04:04:09jhMikeSI figure for voice all that really matters is that it be easily understood
04:04:42Davide-NYCI don't want to get into subjective ramblings, but please define "OK"... ;-)
04:05:31Davide-NYCI'll quickly present a theory. If you can record stuff this 'crappy' we're going to get a lot more GoldenQuotes from all corners.
04:05:32jhMikeSFor voice I don't care about artifacts. Only that it be about as clear as a telephone.
04:05:53Davide-NYCcan't we split the difference? Compromise
04:06:16jhMikeSWell, if some bonehead wants to do DAT archiving at 24kpbs, let 'em.
04:06:31Davide-NYCyeah, you're right
04:06:43Davide-NYCIt's in the statusbar the whole for crying out loud
04:06:48Davide-NYC*whole time
04:07:05jhMikeSIt's just NOTHING is able to be made out at 8kpbs stereo. Maybe if it did joint stereo it would work.
04:07:22jhMikeSThat's for later though
04:07:25Davide-NYCSo the question is from what bitrate will you restrict to mono?
04:07:59jhMikeSAnything under 16 for sure
04:09:00Davide-NYC16k/22kHz/stereo is junk as well, equal voice to artifact (subjectively
04:10:12Davide-NYCHow about all of the exclusively 22kHz settings should be restricted to mono?
04:10:22jhMikeSI may just nix all the 8kbps.
04:10:29Davide-NYCand 16 then!
04:10:35Davide-NYCand 24
04:10:38Davide-NYCstart at 32
04:10:42*jhMikeS is listening to Davide-NYC's voice files
04:10:45Davide-NYCleave everything else as is
04:11:51jhMikeS22k mono is good
04:12:07jhMikeSI mean 16kbps mono
04:12:45jhMikeSand start at 32kbps for stereo
04:13:28Davide-NYCI can agree with that
04:13:35Davide-NYCyou are sage
04:14:23jhMikeSThe voice sounds very good at those rates, better than the answering machine
04:14:46jhMikeSWay better than the drive-through window even
04:15:47Davide-NYCnot familiar with the drive through window (NYC) :-)
04:16:00Davide-NYCNo car even!
04:16:18jhMikeSMaybe actually "driving-through" the window then
04:16:43jhMikeSLess noisy than a drive by
04:17:05Davide-NYCwrite that down son that's bumpin'!
04:17:39Davide-NYCanything you want em to do?
04:17:55Davide-NYCcause I have to call it a day soon
04:18:02 Quit _Veseliq_ (Read error: 110 (Connection timed out))
04:18:10jhMikeShmmm...if it records at all the rates and plays at all the rates we're good.
04:18:30jhMikeSso I can't think of anything else except for SPDIF
04:18:43Davide-NYCI can test that tomorrow or the day after.
04:18:50Davide-NYCI need access to the studio
04:19:07Davide-NYCOtherwise I can only test at 44.1k
04:19:23jhMikeSWould be good enough ATM to see if it even works still
04:19:30Davide-NYCOK I do
04:19:47Davide-NYCdon;t have the right cable (we've been over this)
04:19:57Davide-NYCstill need to go to the studio.'s to get me a cable
04:20:06jhMikeSit's cool
04:20:13Davide-NYCwell they're usually not toslink to toslink
04:20:39jhMikeSI suppose I'll see what I need when my H120 arrives
04:20:56Davide-NYCmaybe they're all called toslink
04:21:18Davide-NYCone has a square outer connector and one is shaped exactly like an 1/8" jack
04:21:21jhMikeSIs it the same connector used for DVD players and such on the H120?
04:21:39Davide-NYCthat's the square end I'm talking about
04:22:01Davide-NYCthe other end that interfaces with the H1x0 is the shape of an 1/8" headphone jack
04:22:10jhMikeSok...can you buy that?
04:22:34jhMikeSspecialty item?
04:22:59Paul_The_NerdPretty common
04:23:31Davide-NYCwhat are they called?
04:23:32Paul_The_NerdThough oddly enough when I tried to buy one at Fry's electronics the guy in the audio cables section was clueless about both the terms Toslink and SPDIF
04:23:34jhMikeSok...some cheap cable that works well is good enough. I've seen some for $90...insane
04:23:45Davide-NYCno no $5
04:23:56jhMikeSabout 5'
04:24:14Paul_The_Nerd5' would probably be in the $10-$20 US range.
04:24:38jhMikeSSaw some on Froogle for like 8.99 but both square ends
04:25:13jhMikeSWhat's the jack type for the H1x0 end called?
04:26:01jhMikeSand the square end is toslink, right?
04:26:22Davide-NYCI presume
04:26:26Paul_The_NerdI think the square end is the regular SPDIF connector and the rounded one is Toslink, but I really dunno
04:26:41Davide-NYCthe other end is shaped exactly like a 3.5mm (or 1/8") headphone jack
04:26:45jhMikeS"Fry's electronics" is a disturbing name
04:27:08scorchefrys is a great store =)
04:27:11Paul_The_NerdIt's pretty much the best one-stop store though
04:27:21Davide-NYCI think SPDIF is the data protocol
04:27:27scorchetheir online store sucks...
04:27:30Paul_The_NerdOnline, I pretty much shop Newegg
04:27:33Davide-NYCyou can have SPDIF over regular RCA jacks
04:27:47Davide-NYCelectrical instead of optical
04:27:48jhMikeSI've got opticals on everything though
04:27:50scorchebut in terms of brick-and-mortar, frys is king
04:27:51Paul_The_NerdI didn't know that digital-coax was still called SPDIF
04:28:07Paul_The_NerdI've always just heard it called 'digital' or 'digital over coax'
04:28:12Davide-NYCI htink it's the transmission protocal much like IP with TCP/IP
04:28:39jhMikeSIt's really EBU
04:28:54Davide-NYCthat, jhMikeS is what you need
04:30:09Davide-NYCto go from one H1x0 to another one would need a dual 3.5mm jack cable
04:30:29jhMikeSpics aren't coming up...hmm
04:31:09jhMikeSI can put two of those together with the TosLinks in my ears
04:31:26 Join |AhIoRoS| [0] (n=ahioros@
04:31:29 Join _Veseliq_ [0] (n=veseliq@
04:31:58Davide-NYCaccording to that auction the jack in question is in fact called TOSLINK and the one on the H1x0 is called "mini-TOSLINK"
04:32:18Davide-NYCso I would need a mini-TOSLINK to mini-TOSLINK cable
04:34:30jhMikeSjust about everything's in the UK for this >:-|
04:35:36Davide-NYCI would lose that converter in a sec
04:35:52Davide-NYCthe 1m cables a safer buy fo rme
04:36:23*jhMikeS would just tape it to the cable real tight
04:37:27Paul_The_NerdI bought some cheap cables, and they've got converters like that tethered to the ends, so it can be TOSLINK->TOSLINK, Mini->toslink or Mini->Mini at my whim
04:38:09jhMikeShmmm...sounds nice...I need two to test feed through from the DVD player to the stereo.
04:38:38 Join Yono [0] (
04:39:12Davide-NYCthat is an elegant solution
04:39:31Davide-NYCall right y'all g'night
04:39:41 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
04:40:30 Quit linuxstb ("CGI:IRC")
04:40:50 Quit midgey34 ()
04:43:31 Join webguest37 [0] (i=43a948b1@gateway/web/cgi-irc/
04:44:23webguest37I am wondering if anyone knows if Rockbox will work on The Archos 404?
04:45:09scorcheit wont
04:45:43webguest37do you know of anything that might work better or i rockbox is going to suppout the archos 404
04:46:35scorchethe 404 is the new one correct?
04:46:50scorchethen i dont know of anything yet
04:46:51webguest37the Ne Gen 4
04:47:03webguest37i got it hooked up for 108
04:47:06scorchebut i doubt rockbox will ever come to it
04:47:58webguest37do you know where i mint find plugins for it as i dn't want to pay 20 bux per plugin to play what i want it to play
04:48:26scorchei dont know
04:48:31scorchethis isnt #archos ;)
04:50:04webguest37just out of curiosity why don't you think rockbox will not come to the new line of archos
04:50:57scorchebecause rockbox (now at least) is meant for mainly audio
04:51:14scorcheit would be quite a waste to put rockbox on that device and waste the screen, etf
04:51:39scorcherockbox is more for DAPs not DMPs
04:52:09webguest37but from what i read it has ogg support and some others
04:52:17scorcheit does
04:52:30scorchewell, i would never get that device for audio playing
04:52:39scorcheand i am not sure who would
04:52:51*scorche shrugs
04:54:47scorcheof course, if someone were to come along who had the capibility to port it and have the device, im sure we wouldnt shoo him away =)
04:54:49webguest37Well see it's not so much i want it for auido i wanted it for video and a bit of auido the problom is it does not support out of the box AAC auido witch alot of my mvoes have
04:55:32Paul_The_NerdOne would think that if your movies were legal, you could encode them with a compatible audio format...
04:55:55webguest37i download them from Usenet
04:56:13webguest37and i havent found a encoder i like have any suggestions>?
04:56:15luckzsee above.
04:56:36Paul_The_NerdMy suggestions are to find another channel, as that's pretty much unrelated to Rockbox, which is primarily and almost entirely audio playback related.
04:57:21webguest37Im just asking some fellow ppl their opinion if you don't want to answer then fucking don't
04:57:29 Quit blue_lizard (Read error: 60 (Operation timed out))
04:57:37 Join blue_lizard [0] (
04:57:51scorchewebguest37: swearing is not appreciated here
04:58:07Paul_The_Nerdwebguest37: This channel has a specific purpose. We like to keep it at or near it.
04:58:19webguest37and being talked to Rudely is not apperciated twords me
04:58:37scorcheno one has spoken to you in a rude way
04:59:59webguest37i asked a question and im tald that apparently no one knows anything about what i am talking about all i need to encode or rather reencode is the auido and i want someones opion on what is a good programto do that
05:00:28Paul_The_NerdYou asked a question and were told this isn't the right place to ask that question.
05:01:18scorchewhich is, of course, true
05:01:50webguest37Alright how about this question AUIDO ONLY ENCODER FROM AAC TO MP3?
05:02:03webguest37What would you two recomend?
05:02:14Paul_The_NerdI'll ignore the fact that it's not actually a question, and suggest dBPowerAmp.
05:02:36scorchebut transcoding is never a good idea
05:02:40 Join linuxstb [0] (i=3f887518@gateway/web/cgi-irc/
05:03:16webguest37Then what about this format if i was going to Transcode auido from aac to mp3 witch would be the best?
05:03:21thegeekI'd think virtualdub could do that
05:03:26thegeektranscode audio only
05:03:37thegeekvirtualdub is a fantastic app;P
05:03:51linuxstbwebguest37: Why transcode? Rockbox can play AAC.
05:04:05webguest37b/c i can not put rockbox on my archos 404
05:04:06thegeekI think the entire point is that rockbox is not supported on his player
05:04:07scorchelinuxstb: PLEASE READ THE LOGS FIRST ;)
05:04:23Paul_The_Nerdlinuxstb: He thinks he's being clever about the fact that we told him discussing video transcoding for his player is off topic for the room.
05:04:45*linuxstb had read the logs
05:05:40webguest37Paul im not anymroe i do have alot of audio id like to transcode
05:06:04 Join webguest12 [0] (i=917405ed@gateway/web/cgi-irc/
05:06:10Paul_The_Nerdwebguest37: There's plenty of places on the internet with better information on it.
05:06:21thegeekare you dyslexic webguest37 ?
05:06:50webguest37or is that on
05:07:04thegeekI'm amazed at your ability to _slaughter_ the english language
05:07:25linuxstbwebguest37: The point is that this channel has a lot of Rockbox development chat, and is logged. So we like to keep talk here on-topic so the logs are not full of irrelevent chat.
05:08:07webguest37Im fucking Russian!
05:08:15myzar|awayyou're fucking a russian?
05:08:16myzar|awaysend pics
05:08:23myzar|awayyou have piqued my interest
05:08:27*Paul_The_Nerd sighs.
05:08:27 Nick myzar|away is now known as myzar (
05:08:38scorchewebguest12: i have told you before about swearing in this channel...
05:08:58*linuxstb sleeps
05:09:01thegeekwebguest37 : english is not my first language either, it's really not much of an excuse
05:09:02 Quit linuxstb ("CGI:IRC")
05:09:06webguest37*Smiles* im innocent like a wee baby goat
05:09:29scorchethegeek: please stop antagonizing him also...
05:09:37Paul_The_Nerdwebguest37: Being foreign is an excuse for grammatical errors, but the sheer number of evident typos where you simple didn't bother to pay attention and transposed letters demonstrates as clearly as your inability to respect the topic your overall disrespect for the channel and its members.
05:09:51thegeekscorche : why, it's soo fun;P
05:10:02Paul_The_Nerdwebguest37: If you need further help with non-Rockbox subjects, I strongly suggest you find other sources of information on the internet.
05:10:19webguest37i feel like staying
05:10:28scorchethegeek: because i feel like i have been thrown into an elementary school ;)
05:10:30webguest37i have growen to like you all
05:11:07webguest37Almost like a family
05:11:08thegeekI'll refrain from replying with the obivious comment ;P
05:13:18 Quit Yono ("Leaving")
05:13:40myzarand i have learned to hate you all, equally
05:13:56myzarand i think of it more like a parasite that eats away at me, day by day
05:14:13myzarbut that's what makes a family a family... right?
05:14:28webguest37Sure does
05:17:03webguest37i mean when you have you mom sucking on you nice hard balls and ou have your sister bent over the arm chair i nthe liveing room...... does anything get better?
05:17:34scorchewebguest37: stop now
05:17:48webguest37come on im just keeping it in the family
05:17:56scorchethats nice
05:18:05scorchebut that isnt near appropriate for this channel
05:18:57webguest37i thought it was to quite in the room
05:19:17scorchethis isnt a social channel
05:19:20webguest37im makeing a movie
05:19:34webguest37Godzilla Mothra porn
05:19:52webguest37it does not get any better than that
05:20:25Paul_The_Nerdwebguest37: Seriously, what is your problem? Is it just fundamentally impossible for you to respect the purpose of the channel?
05:20:43webguest37imagine mathra getting banged aginst buildings screaming out in Exstacy from that nice large godlike godzilla dick
05:20:45webguest37and yes paul
05:21:21webguest37the modeling is going to be a pain in the ass
05:21:29webguest37in mothras ass
05:21:45webguest37ya i wet there
05:22:29webguest37see unlike other ppl i know im an asshole
05:27:26 Quit webguest12 ("CGI:IRC")
05:27:56webguest37no ones has anything else to say?
05:29:16myzaryou scared webguest12 away
05:35:43 Part Paul_The_Nerd
05:42:33***Saving seen data "./dancer.seen"
05:54:00webguest37Nop e in still here
05:54:32webguest37my age? no im 7
05:55:13scorche...and it all comes together
05:55:17 Quit |AhIoRoS| ("Abandonando, see you")
05:56:44webguest37so does any one here want to lend either their artisic tallent to ot voice to my more the godzilla mothra porn?
05:57:16scorchei think everyone in here would rather ignore you
05:57:41webguest37you want to know what i think? i think ppl have already ignored me
05:58:23webguest37besides you cuz our nice kinda like a daddy.... want to fuck me daddy before my real daddy comes home?
05:58:32 Quit webguest37 ("CGI:IRC")
05:58:49scorchebout time...
05:59:15 Join webguest24 [0] (i=43a948b1@gateway/web/cgi-irc/
05:59:34webguest24Hello everyone
06:00:30webguest24:( no one here to answer my questions
06:00:57 Join Paul_the_Ner1 [0] (
06:02:19webguest24I was wondering if Rockbox will ever support the nano?
06:02:37scorche2nd gen?
06:02:51webguest24ya i just got one for my b-day
06:02:53scorcheit will in time...
06:03:00scorchebut it wont be anytime soon
06:03:12webguest24alteast it will support it sometime
06:03:17 Quit _Veseliq_ (Read error: 110 (Connection timed out))
06:06:35webguest24alright guys seriously i sorry for crashing the channel like that im going to go now but be warned i shall return later
06:06:51 Part webguest24
06:06:52 Join InteliWasp [0] (
06:07:26InteliWaspwell i had a good scare from my ipod today...
06:10:00 Join Noah0504 [0] (n=noah@
06:13:44 Join Paul_the_Nerd [0] (
06:14:25 Join _Veseliq_ [0] (n=veseliq@
06:14:38 Part Paul_the_Nerd
06:20:23 Quit pagefault ("Leaving")
06:31:38 Quit Paul_the_Ner1 (Read error: 110 (Connection timed out))
06:37:37 Join Teknomancer [0] (n=Teknoman@
06:38:16 Join goffa [0] (n=goffa@
06:43:48Nimdaeman i wish the video stuff was done for the ipod :(
06:43:51*Nimdae tries to be patient
06:44:07 Quit _Veseliq_ (Read error: 110 (Connection timed out))
06:44:20Noah0504I didn't know there was any video work being done.
06:47:42 Join pagefault [0] (n=pagefaul@pdpc/supporter/active/pagefault)
06:48:07 Join powr-toc [0] (
06:54:33 Join _Veseliq_ [0] (n=veseliq@
06:55:03Nimdaethere's some being done, not much progress yet
06:56:41Noah0504I just hope Theora makes it onto the list of playable video.
06:59:11 Quit goffa ("leaving")
06:59:35 Join goffa [0] (n=goffa@
07:00:08 Join Davide-NYC [0] (
07:05:07 Quit _Veseliq_ (Read error: 148 (No route to host))
07:07:20Davide-NYCjhMikeS: I have a small idea, still around?
07:07:55 Quit Madkiss (Read error: 110 (Connection timed out))
07:09:33 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
07:17:36 Join Quazgaa [0] (
07:23:01 Join Madkiss [0] (i=madkiss@freenode/staff/madkiss)
07:23:20 Join _Veseliq_ [0] (n=veseliq@
07:40:38 Quit pagefault ("Leaving")
07:40:49 Join pagefault [0] (n=pagefaul@pdpc/supporter/active/pagefault)
07:40:58 Quit pagefault (Read error: 104 (Connection reset by peer))
07:41:02 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
07:42:37***Saving seen data "./dancer.seen"
07:44:25 Quit pagefault (Client Quit)
07:45:49 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
07:45:51 Quit pagefault (Client Quit)
07:45:55 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
07:45:59 Quit pagefault (Client Quit)
07:48:59 Join MarcoPolo [0] (
07:49:23 Join mordov|wrk [0] (
07:50:08 Quit _Veseliq_ (Read error: 110 (Connection timed out))
07:52:43 Quit SmilinBob (Read error: 145 (Connection timed out))
07:53:14 Part jhMikeS
07:53:27 Join SmilinBob [0] (
08:00:03 Join pondlife [0] (
08:02:42 Quit powr-toc ("Leaving")
08:03:48 Quit Teknomancer (Read error: 104 (Connection reset by peer))
08:08:28 Join Rob2222 [0] (
08:15:45 Join _Veseliq_ [0] (n=veseliq@
08:22:53pondlifeAny Archos voice users around? amiconn?
08:22:58 Quit InteliWasp ("Leaving")
08:40:55amiconnpondlife: What's up?
08:41:15amiconnI see you removed an mp3_play_stop() in talk.c for hwcodec?
08:42:54amiconnAfaics this shouldn't have been removed
08:44:31amiconnNow shutup() won't work properly on hwcodec for talkbox clips, or when the voice clip is at a frame boundary
08:54:20 Join ender` [0] (i=null@
08:56:25 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:02:58pondlifeamiconn: Sorry, will resolve
09:06:11 Quit EspeonEefi (Read error: 110 (Connection timed out))
09:07:20Nimdaepetur: mahna mhana
09:08:32pondlifeNimdae: : I'd only just removed that from my brain!
09:08:41pondlifeNow I have to start again.
09:08:49Nimdaei'm only here to help
09:09:00scorchedo dooo do do do
09:09:03scorchedo do do do
09:14:09 Quit lostnihilist (Read error: 113 (No route to host))
09:14:40 Join Siku [0] (
09:17:35 Join Teknomancer [0] (n=Teknoman@
09:25:14pondlifeamiconn: Please could you (or another Archos owner) try the new CVS build with voice to ensure it shuts up cleanly.
09:26:06pondlifeI'd also like to get rid of mp3_play_abort() - there's a patch to tidy up the previous patch here: I am just testing on SWCODEC.
09:27:52pondlifeActually forget that patch, just check CVS...
09:28:57 Join Arathis [0] (
09:29:42 Quit idnar (Read error: 145 (Connection timed out))
09:31:10 Join idnar [0] (i=mithrand@unaffiliated/idnar)
09:36:31 Quit Arathis ("[rl_bot quit]")
09:36:44 Quit lini (Read error: 110 (Connection timed out))
09:42:42***Saving seen data "./dancer.seen"
09:44:28 Quit BigBambi ("Leaving")
09:49:51 Join PaulJam_ [0] (
09:53:05 Quit Teknomancer ("Vision[1.2.0-Z-12102006]: i've been blurred!")
09:55:16 Join matsl [0] (
09:59:24 Quit petur ("worrrrk")
09:59:49 Join psiborg_ [0] (
10:00:14 Quit PaulJam (Read error: 110 (Connection timed out))
10:01:06 Join Teknomancer [0] (n=Teknoman@
10:01:20 Quit psiborg_ (Client Quit)
10:06:25 Join leftright [0] (
10:07:38 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
10:08:40leftrightI was listening to Pink FLoyd - The Wall, and I noticed/heard that three track transitions weren't perfectly gapless, there was a slight 'bump' when the tracks changed,
10:10:17pondlifeLinusN: Do you have an Archos handy?
10:10:29LinusNlemme see
10:11:16pondlifeGreat, if you have time could you try applying to CVS and seeing if it still talks properly?
10:12:16pondlifeJust a bit of file/dir spelling, make sure it doesn't confuse the MAS and doesn't cut the clips short.
10:14:10leftrightmy music is encoded with LAME 3.90.3 @aps
10:14:32pondlifeleftright: Did you use −−nogaps?
10:14:56leftright−−alt-preset standard switch
10:15:43pondlifeTry again, adding −−nogap
10:15:52pondlifeDetails are in lame −−longhelp
10:16:10leftrightheh, there is no way i'm going to rerip 120 gigs of music :-)
10:16:19pondlifeJust try "The Wall"
10:16:37pondlifeIt would be good to know that recent buffering mods haven't broken gapless playback!
10:16:51leftrighti'll try when i get home in two weeks time
10:17:00pondlifeOK, I may try it tonight..
10:17:29leftrighti'm on the road right now, no access to misuc collection
10:18:54 Join dpro [0] (
10:22:07LinusNpondlife: next time you commit a patch, please include the patch # in the comment
10:22:33pondlifeWill do, sorry.
10:22:50LinusNwhich one was it?
10:22:55pondlifeIt didn't get onto Flyspray
10:23:13LinusNthat explains it
10:23:39pondlifeStéphane and I have been doing stuff in private, but neither of us has a (working) Archos
10:23:52pondlifeI will insist on it being in public in future.
10:24:27pondlifeAll or my work is..
10:25:51 Join muesli|trillian [0] (i=user@
10:27:24LinusNdon't worry
10:27:40LinusNthe work is public, since it's in cvs
10:28:08pondlifeIf you can confirm that MASCODEC still works, I'll commit #6241 and remove the (IMHO) unneeded mp3_play_abort() interface.
10:28:17pondlifeBefore other stuff relies on it!
10:29:51 Part leftright
10:30:44pondlifeleftright has left, right?
10:30:55LinusNhmmm, the eq settings shouldn't be in the archos zip file...
10:31:07LinusNpondlife: booooooo
10:31:22pondlife;/ I thought it was good for am
10:32:19pondlifeHmm, in metadata.c , the code for AIFF is enabled for MASCODEC. That's not needed is it?
10:32:43pondlifeget_aiff_metadata() and the code that calls it
10:32:59pondlifeI've got a metadata.c update coming up; will change that too...
10:34:05pondlifeNoone was planning on added a MAS codec for AIFF? WAV first I suppose..
10:35:19LinusNi don't know if this is new, but sometimes the menu entries aren't spoken
10:36:00LinusNif i move fast in the menu, it skips some entries, which is ok, but it doesn't play the last one either
10:36:58pondlifeThat might be a new bug. It should definitely skip those as you move but play the last one
10:37:37pondlifeIs that with the patch? If so, does current CVS do that too?
10:37:46LinusNi'll try with cvs too
10:38:04pondlifeOK. And if that's still bugged, maybe yesterday's daily build....?
10:40:58LinusNi can make it happen on current cvs too, but it is a lot harder (only happened once)
10:41:08LinusNso the patch makes it worse
10:42:12pondlifeHmm, back with v2 in a mo..
10:45:24LinusNi really like how the fm recorder doesn't spin up the hdd when you turn it off
10:45:53LinusNi wish the iriver could do the same
10:46:53amiconnpondlife: metadata.c isn't used on archos
10:47:42pondlifeOK, so it shouldn't need #ifdefs for SWCODEC then...
10:47:59pondlifeLinusN: take 2 at if you could do the honours...
10:48:19amiconnNo, currently not.
10:48:21tiaxI made a backup of my ipod 4g grey boot partition which I now want to restore - will dd if=bootpartition.bin of=/dev/sda count=1 suffice?
10:48:24 Join muesliii [0] (i=user@
10:48:45*amiconn hopes that this will change in the not too far future
10:49:01pondlifeamiconn: I'll leave them in then
10:49:05muesliiiamiconn nice work on solitaire btw
10:49:12muesliiilove it :)
10:49:34pondlifeDo we have solitaire on the player yet?
10:50:17muesliiinice game 2 kill time :)
10:50:24pondlifeCharcell doesn't lead to realistic bouncing, sadly
10:50:37scorcheyeah...the bouncing is a must have
10:50:49LinusNpondlife: now it's rock solid, even better than current cvs
10:50:59pondlifeSimpler too
10:51:17*amiconn wants one more feature in solitaire
10:51:27amiconnDon't know how to implement it nicely yet
10:51:29pondlifeI also have a further mod to clear buffered voice (spelling) when playback is started
10:51:34muesliiii have only 3 small wishes for solitaire... 1. playback options like in jewels 2. saving/resume game 3. points!!!
10:51:50LinusNpondlife: spelling works fine too btw
10:52:03pondlifeI don't think anyone enjoys hearing Microsoft Mary talking over the first 10 seconds of their selected song.
10:52:07amiconnI want a button to do all possible moves automatically
10:52:20amiconnWould be nice for the stupid final cleanup
10:52:33muesliiiamiconn yeah..sounds great!
10:52:55muesliiiwot about crtl+alt+del!? :D
10:52:57JdGordonby possible moves do you mean auto- go-to-the-top or all moves?
10:53:24LinusNsolitaire on my tungsten detects that it is solvable and asks if i want to do it manually or automatically
10:53:40amiconnLinusN: Yeah, the idea isn't new
10:53:57muesliiilets have that pls! :)
10:54:11amiconnSoliton on Amiga does have such a button, and it's also the one where I got the auto-unhide feature idea from
10:54:53muesliiihow do u guys think about my wishes?
10:56:41tiaxI tried uninstalling rockbox as described in the docs, yet ipodpatcher.c doesn't compile on linux, so I can't use that command. Will dd do the trick of restoring the original firmware or is there something I'm missing on?
10:58:36 Join spiorf [0] (
10:59:29scorcheit will work provided that you do it right
10:59:38pondlifeIs it possible to detect unsolveable after initial generation and/or every move? Or would that be too slow?
10:59:59pondlifeOr can an unsolvable game be determined from the start?
11:01:00tiaxscorche: so if I just typed dd if=mbr.bin of=/dev/sda count=1 and dd if=bootpartition.bin of=/dev/sda1, would I do it right?
11:03:51scorcheshould only need the second part
11:04:18 Quit muesli|trillian (Read error: 110 (Connection timed out))
11:09:50 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
11:13:15tucozbluebrother, it looks like that problem is only valid for me, as the manuals were built on the rockbox server.
11:13:15scorchetucoz: that reminds me...for the ipods, can you add something in to say that firewire is not supported?
11:13:15tucozbluebrother, however. the only references i found on google to were german pages, and the german manual for makeindex.
11:13:15tucozscorche, sure. do you have an idea where i should add that?
11:13:15scorcheinstallation prerequisites maybe?
11:13:15scorcheunder usb connection
11:13:16 Join ender [0] (i=null@
11:13:25 Quit Landus (Nick collision from services.)
11:13:45tucozI haven't done it yet ;)
11:13:47 Join Landus [0] (
11:14:06scorchei cant thanks you now? =P
11:14:24tucozyour'e welcome :)
11:21:03 Join damaki [0] (
11:22:13JdGordonjust double checking... there isnt any opposition to #ifdefing tagcache in the code is there?
11:25:46tucozscorche, should i add that in the same note as the 'disk mode' note?
11:25:47scorchegood =)
11:26:08 Quit idnar (Remote closed the connection)
11:26:26tucozhmm. maybe a separate note is better.
11:26:32scorchetucoz: i would do it after the first sentence in usb connection, but you are the manual person =)
11:27:48tucozI think people read USB connection, and a note might catch their attention.
11:28:09tucozas opposed to something written in the text.
11:29:17tucozIs there something else i should say, or is it enough to say that Firewire is not supported?
11:29:17 Join idnar [0] (i=mithrand@unaffiliated/idnar)
11:29:30tucozIs it impossible to use rockbox with firewire ipods?
11:29:57scorchewell, firewire will work for charging, but not transfering i believe
11:30:25tucozso it is basically not possible to use it with rockbox?
11:30:41scorcheit causes issues when people do
11:30:54scorcheon the last issue, i checked the manual to see if it was there, and it wasnt
11:30:56amiconnFirewire should work fopr transferring data as well, it's just that rockbox doesn't detect a firewire connection you have to reboot into diskmode manually
11:31:29scorcheamiconn: there was something about the last guy that said he could only transfer 1 gig and nothign past that
11:31:31scorcheit was a bit odd
11:32:21scorcheamong other threads as well
11:32:40scorchebut i guess that thread is just diskmode
11:32:44amiconnIf I had a firewire cable I could even try adding firewire detection
11:33:33amiconn(provided mini G2 supports firewire)
11:35:03tucozis it both usb _and_ firewire in the ipods?
11:35:10tucozi.e. all ipods
11:35:23scorchethe newer ones dont support firewire
11:35:30scorcheso leave out 5g and nano
11:35:49tucoz\note{Firewire detection is not supported in Rockbox at the moment. Please use USB only.}
11:35:57tucozis that enough?
11:36:42scorcheit is for me, but ask amiconn as well =)
11:37:14tucozamiconn, what do you think?
11:37:31 Quit ender` (Read error: 110 (Connection timed out))
11:39:34 Quit idnar (Remote closed the connection)
11:42:47***Saving seen data "./dancer.seen"
11:46:15 Quit tiax (Remote closed the connection)
11:47:23 Join tiax [0] (i=tiax@gateway/gpg-tor/key-0x57BFF984)
11:55:24 Join bbroke [0] (
12:01:26pondlifeKeymapping question on H300 (+ likely H100 and maybe others too)...
12:01:53pondlifeWhen in menus, could Play exit rather than select?
12:02:01 Part muesliii
12:02:36pondlifeThat would give a "return to WPS" from the context (and settings) menus when called from the WPS.
12:02:57Teknomancerpondlife Play DOES go back to WPS no ?
12:03:01pondlifeWe already have both NAVI and right as selects.
12:03:20Teknomancerhuh? here when i press PLAY (when a song is playing) it goes back to WPS
12:03:20pondlifeDuring playback, hold NAVI for the context menu and try it
12:03:37pondlifePlay selects the highlighted menu option
12:03:45Teknomancerok sec
12:03:47Teknomancerso playing a song now
12:03:53pondlifeOnle left gets back to the WPS
12:04:04Teknomancerso now pressing NAVI
12:04:10Teknomancerit takes me to FileView
12:04:14Teknomancernow i press PLAY
12:04:20Teknomancergoes back to WPS
12:04:27Teknomancerit does not select the new song
12:04:37pondlifeHold NAVI
12:05:01Teknomancerso now i'm in General Settings
12:05:16Teknomancersorry in Sound settings
12:05:24Teknomancernow i press PLAY
12:05:28Teknomanceryes goes BACK to WPS
12:05:38TeknomancerH340 EU
12:05:41 Part tucoz ("Leaving")
12:05:45pondlifeMine goess into Volume etc.
12:05:51Teknomancerhere not
12:05:56pondlifeCurrent CVS build
12:06:02Teknomancermaybe we're using different RB builds?
12:06:18pondlifeI'm using the current CVS
12:06:38Teknomancermine is 061010-
12:07:53pondlifeCould you try the latest?
12:08:17Teknomancerpondlife sorry, not right now
12:08:31Teknomancermy USB cable isn't here :( left it at my friends place ... need to get it back
12:08:42pondlifeNo worries
12:08:55pondlifeI just thought it odd behaviour - it stuck out as being wrong
12:09:10pondlifeI'll see if I can work out why it's happening here.
12:10:48PaulJam_for me play in the menu is the same as navi/right
12:10:53PaulJam_on h300
12:11:08TeknomancerPaulJam_ strange, its the same for pondlife as well but not here for me
12:11:19Teknomancerplay always gets back to WPS for me
12:11:27PaulJam_do you use an exprimental build?
12:11:34TeknomancerPaulJam_ yes, KOSH
12:11:36 Join barrywardell [0] (n=barrywar@
12:11:39Teknomancerthe one from MisticRiver
12:12:02PaulJam_it has a patch integrated that changes the behaviour of play in the menus
12:12:13TeknomancerPaulJam_ i see, yeah then that explains it
12:12:18pondlifePaulJam_: 2 questions then!
12:12:35pondlife1) Do you think that this particular use of play should be changed in CVS?
12:12:48pondlife2) Do you still have the white noise problem with recent builds?
12:12:55*Teknomancer goes back to coding in zeta
12:14:01PaulJam_1: what wold the play button do if playback is stopped?
12:14:57pondlifeSame as in the browser? Nothing, or resume last playback.
12:16:18pondlifeI know the arguments about that horrible "left at root goes to WPS" patch, and don't want to subvert anything but this just seems wrong
12:17:11 Join damaki_ [0] (
12:17:26PaulJam_2: i have used a patched build the last few days. with the random folder function the whitenoise happened but only once after the 10th folder or so. previously it happened more often. i haven't tried a build with the commits from today.
12:17:49pondlifeToday's commits shouldn't matter, I was thinking of lostlogic's buffering fixes
12:21:16PaulJam_i could try my test albums with a clean cvs version...
12:21:31 Join TeaSea [0] (n=thunderc@
12:22:29 Quit darkless ("Leaving")
12:22:37 Join darkless [0] (
12:23:20 Quit ender (Read error: 104 (Connection reset by peer))
12:23:25 Join ender` [0] (i=null@
12:23:26barrywardellcan someone advise me on
12:23:54barrywardelli've already fixed the problem another way, but am wondering should the __DATA thing be added in too?
12:32:34 Quit damaki (Read error: 110 (Connection timed out))
12:38:29 Quit Noah0504 (Read error: 104 (Connection reset by peer))
12:40:01 Join _FireFly_ [0] (
12:40:50scorcheamiconn: we have rombox again (without tagcache, but still) =)
12:46:24 Join Wolfsbane [0] (n=Wolfsban@
12:46:32 Quit barrywardell ()
12:50:14Wolfsbanecan any1 send me bootpartition.bin for 1 gen iPod nano please? Formatted it and can't use iTunes due to "software restriction policy"
12:51:11JdGordonits about 80mb.. isnt it?
12:51:50scorche80,284 KB ;)
12:52:40scorchefor mine at least..
12:53:06 Join ender [0] (i=null@
12:53:11JdGordonyeah, a bit big to send :p
12:53:31Wolfsbanetorrent it?
12:53:44 Join MadDog011 [0] (
12:53:49scorchei would rather not upload it to any trackers...
12:53:59JdGordontorrent is useless if only 1 person is seeding
12:54:18 Quit ender` (Read error: 104 (Connection reset by peer))
12:54:18scorchenot necessarily
12:54:23 Part MadDog011
12:54:30JdGordonwell.. you may as well just ftp it
12:54:45scorcheyeah...i wil only send it that way =P
12:59:01 Join damaki__ [0] (
13:02:02 Quit nudelyn ("At Argon, we're working to keep your money.")
13:03:17BigMachey how can I turn of dircache on a 5g ipod
13:03:22Wolfsbanei guess i'll go back badgering our sys people to let me use iTunes then ;-(
13:03:24BigMacI know it is in the manual
13:03:33BigMacbut the site is blocked at my school
13:04:12scorcheWolfsbane: like i said, if you have an ftp, or something like that, i will upload it
13:04:18scorchebut otherwise, it is just too big
13:04:38scorcheBigMac: what is the reason it gives for being blocked?
13:04:41 Join Abst [0] (n=Abst@unaffiliated/abst)
13:04:54BigMacGames and Applications
13:05:05BigMacI know
13:05:07 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
13:05:17BigMacand they blocked out the proxy tab in IE
13:05:30BigMacAnd I don't have my PFF
13:05:41scorchejust look for it in the menus...
13:05:42amiconnBah, bad idea :(
13:05:50BigMacI found it
13:05:55BigMacbut it is greyed out
13:05:55Wolfsbanescorche, you don't mean just an ftp client, do you?
13:06:10scorcheamiconn: ?
13:06:11BigMacyou can't enter any thing into it
13:06:28scorcheWolfsbane: no...there has to be an ftp server
13:06:33BigMacdownload that
13:06:44amiconnI don't want rombox to have a different feature set than plain rockbox
13:06:58scorcheBigMac: that has no relevance...
13:07:20BigMacI thought he misunderstood what an ftp client was
13:07:50LinusNamiconn: then use a very old version of rockbox, problem solved
13:08:03scorcheamiconn: i know, but it is pretty much past the point of just optimizing to get the size close enough
13:08:05BigMacwell I have to get backto class If no one Knows how
13:08:12amiconnscorche: I disagree
13:08:31markunIs it even possible to grey out options in the rockbox menu?
13:08:51 Join kridian [0] (
13:08:52scorcheeven if it is, it isnt like you have to use it
13:09:02BigMacIn IE
13:09:32BigMacI can't just whip out my ipod in class and start looking through the menus
13:10:17scorchewell think...most likely general settings...then it is probably system options
13:10:23scorcheeducated guessing...
13:10:54BigMacWell I'd give myself 15 seconds from taking it out before the teacher notices
13:11:03BigMacso if it is not there
13:11:10BigMacI lost my ipod
13:11:20scorchethen dont use it in school
13:11:23BigMacso i was looking for a more exact answer
13:11:23scorcheproblem solved.
13:11:51BigMacThat is the only reason I have a DAP
13:12:26markunwhy do you want to turn of dircache anyway?
13:13:52 Join idnar [0] (n=mithrand@unaffiliated/idnar)
13:14:36 Quit Abst (Remote closed the connection)
13:14:37preglowi think we pretty much ARE past the point of being able to optimise the size of rockbox down for rombox
13:14:40 Quit damaki_ (Read error: 110 (Connection timed out))
13:14:45preglownot because it's not possible, but because it won't be done
13:16:04BigMacbecause Paul_The_Nerd said it would help improve my batery life markun
13:16:05 Quit _FireFly_ ("Leaving")
13:16:10 Join Abst [0] (n=Abst@unaffiliated/abst)
13:17:30*pondlife takes amiconn's side on this one
13:19:01pondlifeDoesn't JDGordon's big settings patch reduce binary size lots?
13:19:17pondlifeOr am I out of date?
13:19:35 Quit idnar (Remote closed the connection)
13:19:55 Join idnar [0] (i=mithrand@unaffiliated/idnar)
13:19:56pondlifeIf we can add voice support for plugins, then the menus can all go out of the core too.
13:20:02 Quit Rob2222 (Read error: 131 (Connection reset by peer))
13:21:33PaulJam_but wouldn't this mean disk spinup for accessing the menu?
13:22:17pondlifeBut maybe we can only unload a plug-in when the space is needed.
13:22:37PaulJam_btw, on first test with clean cvs on h300 i haven't been able to get whitenoise.
13:22:39pondlifeInitial boot would be quicker, the settings part would only be loaded when needed.
13:23:13pondlifeI'm just saying there's definitely room for optimisation.
13:23:27*petur is against taking settings menu out of the core
13:23:39 Join Genre9mp3 [0] (
13:23:40pondlifepetur: Why?
13:24:17BigMacdoes anyone Know if the logitech bluetooth headphones will work in rockbox
13:24:32 Join _FireFly_ [0] (
13:24:35pondlifeI'd think of tagcache as more core than the settings . That's my opinion.
13:24:41aliaskBigMac: They should, but not the iPod control ones
13:25:28pondlifeIt needn't be any slower unless you've been using another plugin.
13:25:36scorchewell, either way, with HAVE_TAGCACHE not defined for FMrecorder, we have rombox.ucl at 216 KB
13:26:11scorcheso even if it is a temp solution, it will give us some room as well
13:26:38pondlifeGreat. I can boot quickly but not use tagcache. Also that room will probably be wasted now the pressure for optimisation is off a bit.
13:26:59pondlife:( pessimist
13:27:35scorchepondlife: well, like i dont have to use it
13:28:00pondlifeI just think it's the wrong fix for the problem. What do you remove next time you need space?
13:28:38scorchehopefully nothing
13:28:54scorcheeven if it is a wrong fix, it is a fine temp fix
13:29:18pondlifeBetter Rombox that No Rombox, true.
13:29:19SlasheriJdGordon: btw, tagdb/* is not needed
13:29:38SlasheriJdGordon: probably it should be removed
13:29:49 Quit petur ("worrrk")
13:30:08pondlifeJdGordon: Any idea how far from commit the settings patch is, btw?
13:31:00scorcheSlasheri: thanks...i did not realize that it wasnt necessary to not ifdef those out...i figured i should just in case
13:32:04scorchepondlife: well since he doesnt seem to be active atm, i believe he is waiting till after exams to finish it...and there is still the issue with the cpu being boosted
13:32:44LinusNmy opinion regarding rombox is that rockbox will always gain features and will always grow
13:32:56LinusNso we will eventually hit the roof again
13:33:21LinusNyou can only optimize the code so much
13:33:54LinusNand as far as i can see, the size optimizations aren't happening
13:34:09pondlifeI agree with that, but think there is room for core shrinking. We just need to decide what is core and what can be put in a plug-in.
13:34:41LinusNi'm firmly against moving code to plugins
13:34:48pondlifeNot even the settings?
13:34:54LinusNespecially not the settings
13:36:03pondlifeI'm biased I guess. I have CFG files set up, so rarely use the settings menus.
13:36:42pondlifeIt just seems like an seperate application to me. A CFG editor if you will
13:37:40pondlifeObviously I'm not talking about context menus, recording etc..
13:37:47LinusNi would guess that the major part of the settings is not the menu code
13:38:10LinusNbut the code for loading and storing
13:39:10pondlifeThe loading code is the nub here.
13:39:27pondlifeEspecially as we're moving to text storage.
13:39:39pondlifeOr so it seems
13:40:01*pondlife would have a binary image of the settings and just load by reading it.
13:40:12pondlifeSound familiar?
13:40:32LinusN"The word nub is not in the dictionary!"
13:40:43pondlifeI know, but I like it.
13:41:42LinusNbut what does it mean?
13:42:04pondlifeThe main question?
13:42:48***Saving seen data "./dancer.seen"
13:42:56pondlifeI imagine a plugin that deals with the settings and storage to/from CFG files. But the initial settings load would not want to use that.
13:43:10Genre9mp3Hmmm... about the 24bit FLAC commit for the Coldfire targets...
13:43:16LinusNpondlife: exactly
13:43:16Genre9mp3wasn't the UDA incapable of decoding 20+ bit audio?
13:43:40LinusNGenre9mp3: it's incapable of playing and recording 20-bit audio
13:44:00pondlifeSo keep the current settings in binary form for quick startup and minimal memory use.
13:44:14LinusNpondlife: how KISS is that?
13:44:41Genre9mp3LinusN: So the 24bit FLAC commit is only for the X5? (don't know what DAC the X5 has)
13:44:42LinusNhow? you suddenly have two file formats for settings
13:45:02LinusNGenre9mp3: the 24-bit audio is scaled to 16 bits
13:45:05pondlifeWell, no more than we have now.
13:45:15 Quit Quazgaa (Read error: 110 (Connection timed out))
13:45:50pondlifeThe .CFG file viewer is responsible for reading a CFG file and writing the binary image (call it "sector3d", it could be a file).
13:45:50Genre9mp3so coldfire scales it down for the uda to handle it... ok
13:46:04Genre9mp3LinusN: Thanx for the clarification
13:46:13pondlifeVery similar to what we have now, but the CFG handling is out of the core.
13:46:45LinusNpondlife: seems like a pretty complicated way of changing settings
13:46:50 Join Rob2222 [0] (
13:46:52pondlifeSame as now
13:47:06LinusNno, today you just enter the settings menu
13:47:11LinusNand the code is in the core
13:47:59pondlifeWell, the user would still open the settings menu to edit their config (although there may be a spin-up if the plugin isn't loaded).
13:48:16pondlifeApplication code being in the core is not necessarily a good thing
13:48:19LinusNi'm mostly talking about code complexity
13:48:35LinusNpondlife: the application is the core
13:49:11pondlifeI think it simplifies things as the settings code is partitioned off. All the core has to do is read (or re-read) a binary block.
13:49:22 Join My_Sic [0] (i=3ea13b0f@gateway/web/cgi-irc/
13:49:28My_Sichie all
13:49:43My_Sicjust to tell you that with the last cvs build
13:50:23LinusNpondlife: this will make the settings file less compatible
13:50:24My_Sictagnavi_custom.config don't works
13:50:41pondlife(to LinusN)
13:51:05LinusNbecause the binary block will be incompatible as soon as you change the settings struct
13:51:14pondlifeIsn't that currently the case?
13:51:30LinusNonly if you change the size of an existing setting
13:51:34pondlifeAh, ok
13:52:36preglowmoving menus to plugins???
13:52:36pondlifeWe can automate the upgrade process easily though. Just store the name of the last used CFG file in the binary block. Then re-parse it if the version changes.
13:52:37preglowthat's insane
13:52:46pondlifeNot menus. Settings.
13:52:51preglowwell, settings too
13:52:53LinusNpondlife: and that's KISS?
13:53:09pondlifeWell, maybe not that last bit
13:53:18preglowif the purpose is to make rombox fit again, it's all for nothing anyway
13:53:27LinusNand you also get a depencency on the plugin api
13:53:32preglowwe'll hit the limit again and again and again
13:53:37LinusNpreglow: amen
13:53:38preglowwhat those targets need is a fork
13:53:53preglowall the other targets are going to keep seeing new features
13:54:00LinusNand that fork will sacrifice features to fit
13:54:32LinusNor we could stop developing rockbox
13:54:43preglowyes, ofcourse
13:54:47LinusNmost features are "useless" anyway
13:55:26*LinusN still doesn't know what "nub" means
13:56:04SlasheriMy_Sic: what is the problem?
13:56:35pondlifeI didn't want to use the word "core" :)
13:57:17preglowanyway, i'm half-serious with regard to the fork issue
13:57:21*amiconn disagrees about the fork
13:57:27preglowit's the only way to fix the problem
13:57:48amiconnThe real problem isn't rombox, but rather some optimisations not happening
13:57:56My_Sicwith an old version (like 3 days ago) it's working
13:58:00preglowamiconn: and they won't happen either
13:58:02My_Sici have my own menu
13:58:03preglowamiconn: it's part of the problem
13:58:11amiconnThey need to happen
13:58:15preglowpeople do rockbox for fun, and very few people think optimising is fun
13:58:18My_Sicwith the latest version, i have the officiel menu
13:58:43SlasheriMy_Sic: how it doesn't work, is it crashing or something like that? and can you put your custom config file somewhere in the web or dcc to me?
13:58:43 Quit My_Sic ("CGI:IRC (EOF)")
13:59:04pondlifeOptimisations are good for all versions of Rockbox. Things like fast boot-up and large buffer are primary.
13:59:52amiconnamen to that
13:59:53preglowsure they are
13:59:57preglowi agree completely
14:00:00preglowbut i don't think they'll happen
14:00:15preglowi'd be happy to be proved wrong
14:00:15pondlifeI like the Archos for that reason - an encouragement. I would more selfishly fork the MAS though!
14:00:18Mode"#rockbox +o Bagder " by ChanServ (ChanServ@services.)
14:01:01*pondlife smells a new topic cooking..?
14:01:02preglownow, the problem with the archoses is that not enough devs use them anymore
14:01:09preglowso it's not really much of an encouragement for most of us
14:01:18amiconnWell, one of my problems is that I want to see those optimisations happpen, and I certainly want to work on that. But it's demotivating that it looks like I'm almost the only one :/
14:01:22pondlifeIt's an encouragement not to bloat
14:01:32preglowamiconn: like i said, not many people think optimising is fun
14:01:41pondlifeAlways keeping the Archos in the back of your mind when coding....
14:02:04preglowyup, would be nice, but is obviously not happening
14:02:51pondlifeamiconn: I'm up for some optimisation, as long as it doesn't involve assembler.
14:03:03preglowmost good optimisations wouldn't need to involve that
14:03:10pondlifeMy point exactly.
14:03:15 Join Criamos [0] (
14:03:26preglowthe biggest problem is finding spots to optimise
14:03:51preglowmost of the code i work with isn't even used on archos, so i wouldn't know
14:03:57pondlifeWhich is why I mentioned the settings... :/
14:04:09preglowwhat's happening with that patch anyway?
14:04:28pondlifeI'm going to see if I can work out why it's boosting...
14:04:38preglowamiconn: i don't see why only you working on size optimisations is demotivating anyway
14:04:39Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
14:04:45Mode"#rockbox -o scorche " by scorche (
14:04:55 Join My_Sic [0] (i=3ea13b0f@gateway/web/cgi-irc/
14:04:59preglowamiconn: if you start doing it, it might even catch on, heh
14:05:12pondlifeHmm, maybe we need a Feature Freeze?
14:05:22preglowpondlife: we need one, yes, but i don't think it'd work now either
14:05:24LinusNbookmarks.c is a good project to start with
14:05:26pondlifeOptimisation October is almost Over...
14:05:27 Quit TCK (Read error: 145 (Connection timed out))
14:05:37 Join TCK [0] (
14:05:50preglowthe feature freezes i've seen haven't really worked out
14:06:01pondlifeI was joking, should've ;)ed
14:06:15My_Sic<Slasheri> : no crash, no error
14:06:22preglowrockbox is quite unstable these days, so some bug fixing would be really great anyway
14:06:30My_Sic<Slasheri> : i juste only see the official menu
14:06:53pondlifepreglow: Unstable?
14:06:53SlasheriMy_Sic: ok, can you put the file somewhere so i could have a look?
14:06:59My_Sic<Slasheri> it seem that the perso menu file don't loaded
14:07:13My_Sicyes but not now
14:07:26My_Sici haven't my "cable"
14:07:30Slasheriit's probably just a syntax error in the file
14:07:31My_Sicto connect my ipod to the pc
14:07:34Slasherior something like that
14:07:36preglowpondlife: i get quite an amount of lockups
14:07:54My_Sicno beacause it's worrking with an old version (some days ago)
14:08:08My_Sicif i downgrade it's working
14:08:08Slasheriwell, the syntax has been changed a little
14:08:15My_Sicif i upgrad it's donc working
14:08:21pondlifepreglow: Hmm, anything you suspect? Random hangs, I suppose...
14:08:24My_Sicso it could be this
14:08:35My_Sicwhat the change in the syntax ?
14:08:52Slasherinot sure before i see the original file you have
14:08:55preglowpondlife: randomish, i've got a couple of locks when accesing directories and the like, a couple of pdir1full when entering audio debug
14:09:03preglowpondlife: also, usb mode seems to be glitching
14:09:10pondlifepreglow: : Which target?
14:09:13preglowpondlife: h120
14:09:25pondlifeHmm, my H340 seems very happy at the moment.
14:09:42preglowwould anyone know why disk spins up then down before usb screen is displayed?
14:09:52pondlifeDuring playback?
14:09:58preglowi insert usb plug, disk spins up, down, enters usb screen, then spins up again
14:10:02preglowno playback
14:10:09pondlifeJust trying it here...
14:10:16preglowah, hm
14:10:22 Join Athens1 [0] (i=c2db5ccd@gateway/web/cgi-irc/
14:10:22preglowi might just have paused it
14:10:26preglowyou thinking it's saving resume info?
14:10:36PaulJam_preglow: could it le the feature saving the latest changes before usb?
14:10:43preglowpondlife: why would it do that before entering usb?
14:10:49My_Sicok, i send to you my file later (around 15H30)
14:10:52preglowPaulJam_: is that enabled by default???
14:11:31Athens1is it ok to ask a general question here about plugins?
14:11:33preglowamiconn: anyway, you mentioned optimising my flac asm some time ago. can you remember what you were thinking about?
14:11:44preglowPaulJam_: then it's not that, i use almost nothing of rockbox' features anyway
14:11:47pondlifeI see it happening, not sure why
14:11:48preglowAthens1: sure
14:12:04SlasheriMy_Sic: thanks
14:12:15My_Sicthanks to you
14:12:24LinusNpreglow: i don't yet know why it spins down again, i think the ata bridge resets the hdd
14:12:45preglowLinusN: probably wants a known state, yes
14:12:50My_Sicwhere i can find the changing between the old syntax and the new syntax ?
14:13:09preglowi'd like an option to use the eeprom as much as we can
14:13:59Athens1I have updated my build several times in the last month. During this time, all plugins stopped working, and I receive an incompatible version message. Do I need to completely reinstall the bootloader, or is there something easier?
14:14:37PaulJam_pondlife: bad news: i got both, the stopping and the whitenoise with a bleeding edge build.
14:14:47preglowAthens1: how did you update your build?
14:15:16Athens1I unzipped and copied files from recent daily builds
14:15:19pondlifePaulJam_: OK. Random dirs?
14:15:21preglowAthens1: sounds like you have only copied the rockbox.platform file and not the plugins
14:15:31Genre9mp3Athens1: It's surely no need to reinstall the bootloader
14:15:34 Join FuZZy [0] (
14:16:03pondlifePaulJam_: I've been using "Auto Change Dir: Yes" and still haven't heard this.
14:16:11FuZZyhi, i really need some help i've followed teh tutorial on setting up cygwin to compile but i get
14:16:24PaulJam_pondlife: no, the usual procedure: play an album with repeat off, and after it finishes manually select another one.
14:16:27FuZZyarm-elf-gcc:command not found
14:16:39pondlifeLinusN: See what you mean about bookmark.c!
14:17:06FuZZyi dont know whats wrong and i've been searching for an answer but no site seems to be able to tell me
14:17:13LinusNFuZZy: the path to the compiler must be in your PATH variable
14:17:27PaulJam_pondlife: but i had it also with random dir, but this was with a patched build.
14:17:31Athens1Maybe I will delete my plugins before installing the next build so I can make sure that the new build is actually copied. Thanks for the input.
14:17:41pondlifeOK. I'll turn off auto-change dir
14:17:41preglowAthens1: would be wise
14:17:43LinusNFuZZy: is this the one?
14:18:23FuZZynah this one
14:18:45FuZZyo actually i did do it from there
14:18:46PaulJam_pondlife: i think someone said it isn't reproduceable with cbr mp3s.
14:19:01My_SicSlasheri: where i can find the changing between the old syntax and the new syntax ?
14:19:34pondlifeI'm using alt-preset standard on some test albums.
14:19:56amiconnpreglow: Unfortunately not. Do you perhaps remember the timeframe when I said that?
14:19:56LinusNFuZZy: if you did step 3 and step 4, you shouldn't have this problem
14:20:03 Quit Athens1 ("CGI:IRC")
14:21:20FuZZyLinusN: yes i think i assumed it didnt fail in step 4
14:22:22LinusNFuZZy: and you did restart cygwin after step4?
14:22:51pondlifeI wonder if static char current_title[MAX_TAGS][128] could be removed from tagtree.c? 640 bytes is probably worth it.
14:23:37pondlifeAnyway, got to eat...
14:23:46 Part pondlife ("lunch")
14:25:17SlasheriMy_Sic: i don't think there is documentation.. it's not possible to say unless i can see the file and try it if there really is a bug
14:26:15preglowamiconn: nopes, perhaps a month in the past?
14:26:37FuZZylinusN: yah i just did it if u meant opening cygwin again and got lots of commant not found
14:26:42preglowamiconn: i actually think it had to do with what i mentioned yesterday about not always having to read back the residue when adding the result
14:27:36FuZZylinus: i put PATH=/usr/local/bin:/usr/bin:/bin:/usr/X11R6/bin:/opt/sh/bin:/opt/m68k/bin:/opt/arm/bin:$PATH over PATH=/usr/local/bin:/usr/bin:/bin:/usr/X11R6/bin:$PATH in ect/profile and yah it died
14:28:23 Quit My_Sic ("CGI:IRC (EOF)")
14:29:20preglowand i've fixed this where possible
14:30:56 Quit darkless ("Leaving")
14:32:46FuZZywhen i try run configure and selecting normal for the h10 20gb i get arm-elf-gcc:command not found
14:33:16FuZZyeven though i have download arm-elf-gcc just as the rockbox site tutorial said
14:34:53 Join linuxstb [0] (i=3f887518@gateway/web/cgi-irc/
14:35:21linuxstbFuZZy: What gets displayed if you type "echo $PATH" in Cygwin?
14:36:13 Nick myzar is now known as myzar|away (
14:36:53FuZZywait the start up things changed to bash-3.1$ when start up it wasnt like that before
14:37:06FuZZyand it comes up with lots of stuff when i type that
14:37:24FuZZymaybe what ever odd thing has happened will make it worj
14:37:54linuxstbWhich file did you edit to add the path?
14:38:03FuZZyno now commans dont seem to work lol
14:38:42 Join My_Sic [0] (i=3ea13b0f@gateway/web/cgi-irc/
14:39:08linuxstbWhat editor did you use?
14:39:34FuZZywordpad saved it as text doco
14:40:04FuZZybut that came up with no commands stuff so i deleted it and i just realised i forgot to put the copy back but now its running normally
14:40:16linuxstbI expect it converted it to a DOS text file - which probably broke it.
14:40:30FuZZynow when i type echo path i get nothing
14:40:37FuZZyo ok
14:40:45linuxstbDid you keep a backup copy?
14:40:55FuZZyits back to normal now
14:41:19FuZZybut still got the problem of arm error thingy
14:41:49FuZZyi've gor rockbox under rockbox-devel that shouldnt make a diff should it?
14:42:48linuxstbYou still need to edit /etc/profile, just not with Wordpad.
14:43:41LinusNthe odd thing is that the install scripts should have edited /etc/profile and added the path
14:43:54linuxstbOr you can edit with wordpad, but then need to convert it to a Unix text file from within cygwin.
14:44:13FuZZyhow do i do that?
14:44:27blue_lizardor you could use gvim on windows
14:44:27FuZZyor can i just use notbad and ignoe the symbols cause i dont need to see lines
14:44:33amiconnBagder: I have a question regarding Makefiles.
14:44:42blue_lizardand convert filetype by using :set ff=unix
14:44:47LinusNor is it perhaps wordpad that saves it as /etc/profile.txt ?
14:45:07linuxstbHopefully not even wordpad would change the filename...
14:45:07amiconnIiuc, it is necessary to 'export' a variable in order to make it available in sub-makefiles
14:45:14FuZZyyah it did i changed it
14:45:16FuZZystill didnt work
14:45:34amiconnHowever, I wonder how this works for $(SILENT), as that doesn't use export in the toplevel Makefile...
14:45:54LinusNFuZZy: ls /opt/arm/bin
14:46:12Bagderamiconn: SILENT is set by tools/
14:46:19Bagderwhich is included
14:46:51FuZZyi get a lot of arm-elf files
14:47:05amiconnOh? The toplevel makefile also set $(SILENT) ...
14:47:20Bagderyeah, it gets set on several places
14:47:27*amiconn needs to know where to put the info function
14:48:06amiconnDo we want the info texts (like "LD solitaire.rock") to appear even in verbose mode?
14:48:45Bagderamiconn: it could possibly make it easier for people to know what's actually done when "LD..." etc is mentioned
14:48:47LinusNFuZZy: how does /etc/bash.bashrc look?
14:49:17 Join darkless [0] (
14:49:34FuZZyall commented
14:49:48amiconnI also need a suitable name for that function. It will handle both the info text printing and the $(SILENT) bit (in order to save on line length for readability)
14:49:59amiconnI just used PRINT in my tests
14:50:35 Join lee-qid [0] (
14:50:58FuZZyu're right it does change it to dos text not lin i saved in wordpad and opened it in notepad it apears normal now
14:51:15FuZZyso how do i change crlf to lf?
14:51:58blue_lizardbest is to use an editor that can do it himself
14:52:08blue_lizardjedit, gvim and such
14:52:10linuxstbFuZZy: Something like: "tr -d '\015' < /etc/profile > a" followed by "mv a /etc/profile"
14:53:24linuxstbBut as blue_lizard said, if you're planning on doing anything with Rockbox, you'll need a better editor - one that preserves the line-endings.
14:53:40 Join Neko__ [0] (
14:53:41 Nick Neko__ is now known as kridian_ (
14:53:48FuZZyyah, do u know of any good ones
14:53:57 Quit kridian (Nick collision from services.)
14:54:03 Nick kridian_ is now known as kridian (
14:54:45LinusNFuZZy: "all commented"?
14:54:59linuxstbI think jedit, textpad and notepad2 are common ones.
14:56:05FuZZycool thanx
14:56:38blue_lizardi want to mention gvim again
14:56:52blue_lizardvery fast on windows
14:57:29*LinusN wants to be an iKong :-)
14:57:31FuZZyk will give gvim a go
14:57:55blue_lizardbut of course gvim ist definitively not for an beninner
14:58:48blue_lizarda link for people that want to learn vim
14:58:53 Join Rondom [0] (
14:58:58blue_lizardnot my page dont wonder
14:59:17*linuxstb spots some vi evangelism going on...
14:59:27 Join CriamosAndy [0] (
14:59:58blue_lizardnot vi
15:00:14blue_lizardit is old, it is outdated it is dead
15:00:15FuZZyok thanx for all the help blue lizard and linuxstb
15:03:44 Quit My_Sic ("CGI:IRC")
15:03:45 Join My_Sic [0] (i=3ea13b0f@gateway/web/cgi-irc/
15:06:35 Quit My_Sic (Client Quit)
15:06:40 Join My_Sic [0] (i=3ea13b0f@gateway/web/cgi-irc/
15:07:16 Join Rincewind [0] (
15:10:21 Join Nico_P [0] (
15:10:32 Quit FuZZy ()
15:11:43 Join Kitt0s [0] (
15:15:24 Quit Criamos (Read error: 110 (Connection timed out))
15:16:30 Quit My_Sic ("CGI:IRC")
15:17:20 Quit Rincewind ("Bye")
15:17:40 Quit SmilinBob ("Leaving")
15:19:37 Quit Nico_P (Read error: 113 (No route to host))
15:19:38 Join damaki [0] (
15:20:39 Quit damaki__ (Read error: 104 (Connection reset by peer))
15:22:33 Quit mordov|wrk ("The amount of experience you have is paralell with the amount of equipment you have ruined.")
15:23:40 Join pondlife [0] (
15:24:08Kitt0sLinusN, you there?
15:25:29Kitt0scan you mabye help me with trying to complie the dual x5 bootloader with the one i made?
15:25:39Kitt0sspeaking of this patch:
15:26:27LinusNi have never built the dual bootloader
15:30:17 Quit linuxstb ("CGI:IRC (Ping timeout)")
15:32:13 Part LinusN
15:34:09 Quit Febs (Read error: 110 (Connection timed out))
15:36:21 Join funky [0] (
15:42:24lostlogicpondlife: any new info on what I've gone done broken with crossfade and/or skipping tracks? :)
15:42:50***Saving seen data "./dancer.seen"
15:44:24pondlifeNo, sorry
15:44:35pondlifeI've not had time to play
15:44:38lostlogicnp, clearly it's not _that_ annoying the n:_D
15:44:54pondlifeIt doesn't seem much worse than before to me.
15:45:28pondlifeLike I said, I think you may have just modified timings so the underlying flakyness of crossfade (which is easy to repro, btw) is showing more.
15:45:38lostlogicok −− I wonder if it has something to do with what lowlight had seen in the simulator with the request_failed coming back before the check_new_track _apparently_ executed.
15:46:10pondlifeI didn't see that, when was it reported?
15:46:21 Join pixelma [0] (i=pixelma@
15:47:22 Quit aliask ("Chatzilla 0.9.75 [Firefox]")
15:47:49lostlogicnot sure, a couple of days ago −− if synchronization is happening, it should be impossible, but that's what he mentioned
15:51:05 Join lowlight [0] (i=c730190a@gateway/web/cgi-irc/
15:51:54lostlogicspeak of the devil
15:52:12lostlogiclowlight: you figure out what was going on with that codec_request_failed coming back before hte check new track?
15:53:40lowlightlostlogic: I put in a check for queue_empty(&codec_callback_queue) in codec_load_next_track() and it's coming back false
15:54:17*JdGordon returns... 3 hours after being pinged :p
15:54:50lowlightthe queue still has Q_CODEC_REQUEST_FAILED in it
15:55:06 Part tiax ("leaving")
15:55:30lostlogicah, so something is returning before it should, leaving the queue in a Bad State (TM)
15:55:37pondlifeHmm. is there a logf around which includes the logf queueing info?
15:55:54JdGordon[21:30] <pondlife> JdGordon: Any idea how far from commit the settings patch is, btw? <- after exams ill get back into it (so about 2.5 weeks)
15:55:56lostlogicbecause all of those callbacks between codec and audio thread should be lock-stepped by the queue system.
15:56:18pondlifeJdGordon: Sorry, didn't mean to pester. I think I know now...
15:56:28JdGordonno problem :)
15:57:08lowlightpondlife: I still can't reproduce the stopping bug with logf enabled.
15:57:44pondlifeAh, so this queuing thing is not related to that? I assume you had logging to find that out.
15:58:41lowlightI've put in selective debugf's
15:59:38pondlifeOK, so we know that the stop occurs when the queuing issue occurs?
15:59:53pondlifeThis is with VBR MP3s?
15:59:59pondlifeSo many questions....
16:00:27lowlightThis is the "play a directory, let it stop, then play another directory and playback stops after the first track" bug
16:01:41pondlifeDo we know which request failed? I assume that the "normal" end of playback is leaving that in the queue.
16:01:44lowlightseems like something going wrong in the codec_callback_queue when the first directory stops because there's still a request_failed in the queue when it gets to the check_new_track.
16:02:11 Quit PaulJam_ (Read error: 110 (Connection timed out))
16:02:57 Join My_Sic [0] (i=51410ec9@gateway/web/cgi-irc/
16:03:05lowlightBut I can't figure out why it doesn't always occur...what's the connection with logf?...why didn't I see the problem with cbr mp3s?
16:03:07 Quit kridian ()
16:03:26pondlifeMust be a very thin slice of time involved somewhere
16:05:45lowlightpondlife: right...maybe it's a scheduler thing where that thread is put to sleep.
16:09:27 Join Larsie [0] (i=53538563@gateway/web/cgi-irc/
16:09:29*lowlight is getting tired of listening the same album over & over & over again
16:09:35*pondlife too
16:09:56 Quit NickDe (Remote closed the connection)
16:09:58markunhi Larsie
16:11:29Larsiemaybe I'm going to give a try on doing usefull things for rockbox tonight or something
16:11:35Larsieit's hard :P
16:11:57Larsiehard to learn and keep on learning
16:13:05Larsieis the button layout already finished for plugins etc. for the sansa?
16:13:55BagderLarsie: nope
16:14:08Bagderbarry has some changes to commit I believe
16:14:12preglowBagder: mrh has hardware now? woot!
16:14:24preglowthan he unstoppable!
16:14:28preglowis too!
16:15:12Kitt0s!seen RaeNye
16:15:51My_SicSlasheri: hie again
16:15:52LarsieOk, so I have to wait untill barry commits the changes untill now to see what needs to be done :P
16:16:07BagderLarsie: yes, I think that makes sense
16:16:26My_SicSlasheri: here my custom config for the tagmenu :
16:16:30Bagderunless you feel like making an ATA interface for accessing the NAND flash ;-)
16:16:53Larsiehehe, I don't think I'm that far yet :P
16:19:29pondlifeAnyone know why the recording screen turns off voice menus? Would be nice if it always turned it back on.
16:19:40pondlifeBetter if it didn't have to though.
16:21:24SlasheriMy_Sic: ok, lets see :)
16:21:27JdGordonapps/tagdb/ is really a seperate app and not part of rb right?
16:21:41pondlifeI thought it was obsolete now
16:21:54Bagderit is
16:22:11JdGordonso, do i need to undo my last commit on those files?
16:22:23Bagderwe should remove that dir and its contents
16:23:47JdGordondoes that should mean it will be done?
16:24:07pondlifeWho knows about recording, apart from jhMikeS, petur and mmmm ?
16:24:37JdGordonto remove the dir remove it locally then do cvs remove apps/tagdb then commit right?
16:25:35 Join FuZZy [0] (
16:26:30FuZZyhi, when i made cygwin my xp account was "andrew green" i think the spac is screwing it up as whne i go make it says no file
16:27:35JdGordonBagder: should i remove it? or leave it for someone else to?
16:29:02preglowpondlife: i know a little
16:29:12BagderJdGordon: I say remove it
16:30:46SlasheriMy_Sic: just a moment, i need to debug that a bit
16:30:52JdGordonboom! goned
16:31:01SlasheriMy_Sic: but at least you should rename your menu to "custom" or anything other than main
16:31:08pondlifepreglow: Do you know why recording.c disables voice menus?
16:31:16Slasherior it will overlap with the "main" menu in tagnavi.config
16:31:36pondlifeI have commented out that line and all seems to work ok with no unwanted speech...
16:31:41preglowpondlife: might be because you can't play voice and record at the same time on archos
16:31:54preglowpondlife: doesn't explain why it does so on swcodec, of course...
16:32:02pondlifeVoice has already been turned off.
16:32:10pondlifeThis is just a mod to global_settings.
16:32:18My_SicSlasheri: if i rename to custom, i will have the official menu, with an option custom
16:32:36pondlifeMaybe this is an old fix. The comment reads /* recording_menu gets messed up: so reset talk_menu */
16:32:41My_Sicit's allow to have your own root menu
16:33:06SlasheriMy_Sic: you can say %root_menu "custom"
16:33:09Slasherithat should do it
16:33:18pondlifeBut the recording menu still seems to work fine.
16:33:19 Join linuxstb [0] (i=3f887518@gateway/web/cgi-irc/
16:33:26My_SicSlasheri: If there are multiple root menu line, the first one is used
16:33:34SlasheriMy_Sic: yep, correct
16:33:42FuZZynever mind found it in ect/passwd just changed all the names
16:34:08My_Sici will try
16:34:08preglowdo we have a computer app that can make tagcache databases?
16:34:13pondlifeNever mind, I think I know...
16:35:02 Quit damaki (Read error: 104 (Connection reset by peer))
16:35:14Slasheripreglow: yes, the perl script..
16:35:16 Join damaki [0] (
16:35:22My_SicSlasheri: don't work
16:35:27Slasheribut afaik, that is limited to mp3 files only
16:35:29preglowSlasheri: right, but still nothing based on rockbox source code?
16:35:36SlasheriMy_Sic: yes, i know. still debugging
16:35:37FuZZyjust out of curiosity with the h10 and other models with the touch pads, is it easy to get acces to the different pressure points
16:35:38linuxstbPlus I think someone wrote an itunes->tagcache converter - it's in the forums, and maybe the wiki.
16:35:45Slasheripreglow: nope
16:36:08 Join PaulJam [0] (
16:36:44linuxstbA generator based on the Rockbox source could be useful for finding the bugs that cause tagcache building to fail.
16:38:23 Quit FuZZy ()
16:38:43 Quit _FireFly_ ("Leaving")
16:43:55 Join damaki_ [0] (
16:44:52 Join Ainulindale [0] (
16:45:58markunhi Ainulindale
16:46:08AinulindaleIs anybody able to tell me what "10%" implies for the power management of the iAudio X5 ? :-)
16:46:31AinulindaleBecause I just installed and choked on "5h left".
16:46:43Bagderit implies that the page hasn't been updated in ages
16:46:52AinulindaleOh :-)
16:47:18Bagderand the estimation is not calibrated properly, so don't believe it
16:47:26AinulindaleThat was my guess.
16:47:36AinulindaleIs there any mean to calibrate it ,
16:48:05preglowi'm absolutely certain i've read the datasheet of the ata -> flash chip in the nano, but i can't find it now
16:48:12BagderAinulindale: yes, by measuring the power consumption of a running player in a few states
16:48:37Ainulindale(Sorry for my poor english, I'm french, and at work, so I'm not at all trying to be as correct as I would normally be trying to)
16:48:44markunAinulindale: but make sure to update to the latest version because there was a huge improvement last night
16:49:00AinulindaleI just installed it two hours ago.
16:49:08AinulindaleNot the CVS one though
16:49:27AinulindaleBut I have to say I'm quite in love.
16:49:36Bagderthe fix was made after this morning's daily
16:49:44Bagderso tomorrow's version will be a lot better
16:49:51preglowhahaha, the ipodlinux people have identified the ata -> flash chip as the firmware flash
16:50:04markunAinulindale: with the cvs build it was reported that MP3 playback dropped from 100% CPU boost to 0%!
16:50:27lowlightpondlife: what does audio_track_count() return? # of tracks left on the buffer?
16:50:29Ainulindalemarkun: Ok. Whose feet am I supposed to lick now ?
16:50:40AinulindaleOr which deity am I suppose to worship ? :-)
16:51:18AinulindaleOk. Will do. :-)
16:51:23 Join damaki__ [0] (
16:51:29markunI would really like to see a runtime test with a X5L now :)
16:51:37*Bagder agrees
16:51:54 Quit Teknomancer ("Vision[1.2.0-Z-12102006]: i've been blurred!")
16:53:02markunIs there no iaudio runtime wiki page yet?
16:54:02Bagderhere's a fun thread:
16:54:11pondlifelowlight: audio_track_count() is meant to be the number of buffered or part-buffered tracks, I think. I'd like to remove it though, as part-buffered seems a bit useless
16:54:50pondlifeIt's the number displayed on the audio thread debug screen.
16:55:46pondlifeDo you think it might be safe to remove the block in audio_check_new_track() that uses it?
16:56:22pondlifeI would think the memory buffer check should catch that case anyway.
16:56:56pondlifeBut I probably don't understand the full implications of the comment: /* If it is not safe to even skip this many track entries */
16:57:09lowlightpondlife: something seems wrong...when i print that track_count at the beginning of audio_check_new_track(), it says track_count=2 when playback stops at the end of my first directory.
16:57:32pondlifeThis is "normal"
16:57:51pondlifeAs in, it happens here, every time I think.
16:58:37pondlifeJust play a single track and it starts at 2, then doesn't change
16:58:50 Quit damaki (Read error: 110 (Connection timed out))
16:59:02 Quit PaulJam (".")
16:59:05pondlifeIt's definitely not a useful statistic
16:59:31pondlifeWhen you play the second album, trrack_widx and track_ridx are zeroed anyway
17:00:37lowlightyeah...I see the calculation...track_widx is valid even if there's no track there
17:00:47pondlifeI'm not sure why we zero the track and buffer indices, I would think they could be zeroed in audio_preinit, then allowed to run free and wrap naturally.
17:01:02pondlifeSlight increased chance of a small disk read at end of buffer I suppose.
17:02:38 Quit bbroke ("Chatzilla 0.9.75 [Firefox]")
17:03:22SlasheriMy_Sic: found it, there was a buffer overflow in the code
17:03:26SlasheriMy_Sic: testing now..
17:05:24 Quit damaki_ (Read error: 110 (Connection timed out))
17:08:30preglowamiconn: there?
17:09:09 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
17:11:51 Join MarcoPolo [0] (
17:12:31preglowamiconn: it seems it's the soft reset that kills the nano, not the sleep itself
17:13:00 Quit pondlife ()
17:13:08daurnwhats a 2 letter abriviation for stop in a media player? (eg, play is |>, pause is || etc)
17:13:55Kasperlehow about "[]"?
17:14:05preglowdaurn: why'd you want that?
17:14:24SlasheriMy_Sic: a fix has been committed now
17:17:07daurnthx Kasperle
17:17:18daurnpreglow: making web interface
17:17:20daurnneed alts 2 letters long
17:17:23Kitt0swich one should i use ?:\
17:19:10 Quit JdGordon ("Konversation terminated!")
17:19:50My_Sicwhere i can find the fix ?
17:21:18SlasheriMy_Sic: download the latest bleeding edge build (as soon as it has finished compiling)
17:28:41Kitt0sdoes any1 here use iPod Photo?
17:29:27 Quit lowlight ("CGI:IRC (EOF)")
17:30:00amiconnBagder: Any opinion on the function name (info in Makefiles)?
17:32:13preglowamiconn: any idea about soft reset troubles?
17:32:32preglowamiconn: i've now only ifdeffed out the soft_reset code, and it works perfectly
17:32:37 Join Paul_the_Nerd [0] (
17:33:17amiconnI don't know. Soft reset is part of the ata standard.
17:33:35preglowprobably the wait for rdy part that hangs
17:33:38amiconnDo you have the datasheet for the ata flash chip?
17:34:24Kitt0sdoes ipod photo still have Frequency Scaling?
17:34:40preglowamiconn: ipod has a "k" variant of this chip, but i can't find a datasheet for that
17:34:57preglowcan someone tell me why the flaming hell the backlight sometimes never gets turned off?
17:36:04 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
17:39:29*linuxstb spots that winzip 11 beta contains compression for wav files.
17:39:34preglowlinuxstb: wavpack
17:40:22linuxstbIs it?
17:40:26Kitt0sdoes ipod photo still have Frequency Scaling problem?
17:41:15preglowlinuxstb: i think it is
17:41:20preglowlinuxstb: i remember reading it some place
17:41:21linuxstbKitt0s: No changes have been made to CVS, but there are some patches which seem to work. Hopefully a fix will be in CVS soon.
17:42:27Paul_the_NerdWow, I like how it's listed as a "new feature" with Winzip 11 that you can rename the extension on .zip files.
17:42:38Paul_the_Nerd"Definable Alternate Extensions"
17:42:51***Saving seen data "./dancer.seen"
17:42:56linuxstblowlight: Yes, I found that... Seems David Bryant wasn't aware of it.
17:44:09amiconnpreglow: I can't see how the soft reset could hang completely. It should time out...
17:44:18lowlightoh....and new vorbis AOTUV out too
17:46:18preglowamiconn: the timeout is vERy long, looks like 80 seconds
17:47:23preglowhmm, if i change retry_count to 0, it hangs for about 5 seconds, then gives up
17:47:41preglowlooks to me like wait_for_rdy waits ten seconds at a time, though
17:48:21 Join webguest69 [0] (i=c9e0c002@gateway/web/cgi-irc/
17:54:01My_SicSlasheri: thank you very much. it's working
17:54:31amiconnpreglow: "Recovery from sleep mode is accomplished by simply issuing another command (a reset is not required)."
17:54:58amiconnLooks like we don't need the soft reset for wake-up on the nano
17:55:31amiconnWe don't even need to send the sleep or standby command, as the chip goes to sleep automatically with a very short timeout (15 ms)
17:56:19preglowamiconn: i know, do you think i should keep the code in or ifdef it out? i'm somewhat in favour of keeping the number of ifdefs down
17:56:42preglowamiconn: as of right now, the only ifdef i have in ata.c is in perform_soft_reset
17:56:54 Quit MarcoPolo (Remote closed the connection)
17:57:09amiconnWe should at least fake enough of the sleep logic to make code relying on it work
17:57:34amiconn(battery_bench, possibly other code monitoring sleep state)
17:57:40preglowamiconn: why fake it? if i leave the code exactly as is, but just fake doing the soft reset, shouldn't the rest of the sleep logic work just fine?
17:57:58amiconnSure it should
17:58:17amiconnI'm not a friend of obsolete code though
17:58:53amiconnThe strange thing is that the soft reset breaks on nano. The datasheet explicitly mentions it...
17:59:46AinulindaleHmmm, is there any benchmark for rockbox on iAudio X5 ?
18:00:21AinulindaleAnd I don't mean benchmarks from <insert a long time period> ago
18:00:27 Join MarcoPolo [0] (
18:00:41AinulindaleMarcoPolo: Supelec is evil :->
18:01:05MarcoPoloAinulindale: where are you from ? :)
18:01:29AinulindaleMarcoPolo: Paris. ENST :-)
18:01:37MarcoPoloAinulindale: je vois ça :p
18:02:02MarcoPoloAinulindale: you develop for rockbox?
18:03:12preglowamiconn: weird indeed, but btw, shouldn't wait_for_rdy block for ten seconds?
18:03:12 Quit lowlight ("CGI:IRC (EOF)")
18:03:23 Quit spiorf (Remote closed the connection)
18:04:13preglowamiconn: i'm not a fan of obsolete code either, but i'm also not a fan of ifdefs, and i think the nano build can withstand a couple of hundred bytes of unused code
18:04:51AinulindaleMarcoPolo: Nope.
18:04:53AinulindaleA mere user.
18:05:09amiconnwait_for_rdy should block for 10 seconds in the worst case
18:05:50preglowand it looks like it blocks for five
18:05:55preglowi try a 'browse plugins'
18:05:59preglowthen it hangs for five seconds
18:06:01preglowthen nothing happens
18:06:25preglowamiconn: also, the comment in perform_soft_reset mentions 30 seconds, but 80(!) is done
18:06:55 Join bluebrother [0] (
18:09:47My_SicSlasheri: an other problem with tagnavi : when you select "All tracks" in the menu they keep the fmt_identifier until take the new one.
18:11:19 Join _FireFly_ [0] (
18:11:20amiconnpreglow: Yes, I also wondered about that.
18:12:35amiconnWow, that comment is from 2002
18:13:29amiconn...and back then, the timeout in wait_for_rdy was 4 seconds
18:19:23 Join Febs [0] (i=2662c44b@gateway/web/cgi-irc/
18:22:06 Nick ender is now known as ender` (i=null@
18:22:32 Quit linuxstb ("CGI:IRC (Ping timeout)")
18:28:18FebsInteresting article regarding Apple DRM:
18:29:26Paul_the_NerdIf he does pass the legal hurdles, does that mean we could probably use his old PlayFair code so that Rockbox could actually mirror *all* the features of Retail iPod?
18:29:46preglowamiconn: should we lower the amount of retries again, then?
18:29:51 Join qailer [0] (
18:29:56preglow80 seconds is a bit excessive
18:34:14 Join random81 [0] (
18:36:59 Part Paul_the_Nerd
18:37:04Genre9mp3Febs: Eeerr... how can he possibly managed to crack the encrypted audio stream? Maybe he just managed to unlock the files by using the user keys?
18:37:54 Join lightyear [0] (
18:39:47 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
18:42:53Genre9mp3Febs: nevermind, according to Wikipedia he managed to do the second
18:43:28preglowhaven't real already done stuff like that?
18:45:05 Join [TCK] [0] (
18:45:58 Quit [TCK] (Client Quit)
18:46:14 Quit matsl (Read error: 110 (Connection timed out))
18:48:31hcspreglow: I think so, but DVD Jon is more likely to release source
18:49:01preglowhcs: it says he won't...
18:49:10hcsbah, we'll see
18:49:10preglowat least i assume he won't
18:49:22HClhe plans to sell them.
18:49:35HCland license the source to companies
18:50:34 Quit TCK (Read error: 148 (No route to host))
18:51:05 Join TCK [0] (
18:51:12lowlightlostlogic: regarding the stopping bug from earlier...looks like something is happening in the audio thread/queue when stopping at the end of a directory...
18:51:19Genre9mp3$$$ vs. ideology: 1 - 0 ... welcome to the real world
18:52:27lowlightthere are successive Q_AUDIO_CHECK_NEW_TRACK events until codec_load_next_track() finally exits and the codec stops.
18:53:03hcsUSEC_TIMER is something specific to the portalplayer, right?
18:53:28 Quit TCK (Client Quit)
18:53:32lowlightbut that fills the codec_callback_queue with several Q_CODEC_REQUEST_FAILED's
18:53:43lowlightcausing playback to stop after the next track is played.
18:54:59hcsIs there an equivalent mechanism on the coldfire?
18:55:28preglowamiconn: the hang i experience when setting retry_count = 0 is definitely 5 seconds
18:55:57 Join TCK [0] (
18:56:16 Quit JoeBorn ("Leaving")
18:57:04 Quit damaki__ (Read error: 110 (Connection timed out))
18:57:08 Join Id2ndR [0] (
18:57:49 Quit lowlight ("CGI:IRC (EOF)")
18:58:21 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
18:59:27hcsmaybe a cycle counter somewhere?
18:59:43preglowamiconn: actually, it seems to be five seconds no matter what... which suggests that the rdy waiting isn't the problem
19:01:21preglownah, that's not it at all, i removed the entire 30 sec wait loop and it still delays the same time
19:01:24 Join GreyFoux [0] (
19:06:15 Quit Larsie ("CGI:IRC")
19:06:29preglowahhh, the read timeout is five secs
19:11:31 Quit bawb2 ("seacrest out!")
19:12:11 Join rretzbach [0] (
19:17:46amiconnpreglow: Hmm, there are differences between the ata spec and the ata flash chip spec
19:18:16 Join bawb2 [0] (n=bawb2@
19:18:22amiconnWhen a standard ata device is put into sleep mode, it can only be woken up again by a soft or hard reset
19:18:44amiconnThe flash chip automatically wakes up from the next command
19:22:43preglowamiconn: the differences i've found shouldn't really matter with a correctly coded driver
19:22:55preglowamiconn: it shouldn't really care if it gets a soft reset anyway
19:23:08amiconnYes, but for some reason it does...
19:24:19preglowbut yes, it seems the soft reset itself seems to work
19:24:21 Quit Nimdae (Read error: 104 (Connection reset by peer))
19:24:25preglowbut the next read that's attempted times out
19:26:45preglowamiconn: anyway, battery_bench seems to work
19:28:22 Join GFoux [0] (
19:29:00 Join dan_a [0] (n=dan-mirc@
19:30:17 Join Nimdae [0] (
19:32:56 Part pixelma
19:34:47 Quit edx_ ()
19:37:31 Join petur [0] (n=petur@rockbox/developer/petur)
19:42:52***Saving seen data "./dancer.seen"
19:42:58 Join TeaSeaLancs [0] (n=thunderc@
19:45:51 Quit GreyFoux (Read error: 113 (No route to host))
19:48:06 Join bundik [0] (
19:49:20 Quit akaias (Read error: 60 (Operation timed out))
19:51:21 Join akaias [0] (n=akaias@
19:54:17 Quit Abst (Remote closed the connection)
19:54:22 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
19:54:29tucozbluebrother, hi
19:56:06amiconnHmm, there's an odd effect in the id3 viewer, resulting from it marking the current entry and its corresponding 'identifier':
19:56:39amiconnOn the player, and on all other targets when using the arrow cursor, there are 2 arrow cursors in front of the lines
19:56:54amiconnEspecially odd on the player as there are only 2 lines in total
19:57:31tucozbluebrother, did you read about my problems with the file?
19:57:43 Quit TeaSea (Read error: 110 (Connection timed out))
19:58:01preglowlet's try abolishing the arrow curser again
19:58:09preglowit's ugly and smelly
19:58:19amiconnNot possible on player though
19:58:46preglownono, but everything else
19:58:54*amiconn votes for ditching the arrow cursor on bitmap targets
19:59:06tucozgets my vote as well
19:59:07preglowi just remember what happened the last time linus tried it...
19:59:07amiconnWould even save some code
19:59:13amiconnyeah :/
19:59:28tucozwe need a bdfl on this project
19:59:36*preglow volunteers :P
19:59:46preglowbenevolent dictator for life, afaik
20:00:00amiconnoh noes! ;)
20:00:27tucozsome projects have one. Like ubuntu for instance
20:00:45tucozand I think Linus Torvalds is one for the Linux kernel
20:00:51*goffa raises hand for the bdfl :)
20:01:33 Join charkins [0] (n=casey@pdpc/supporter/active/charkins)
20:01:44Galoisisn't hans reiser also a BFDL
20:01:53*lowlight votes for removing the arrow cursor
20:01:59preglowGalois: he's benevolent, for sure
20:02:00goffaum... no he's a mfdl now :)
20:02:08amiconnpreglow: Do I remember correctly that the pp gpio has this bitmasking stuff for selecting what bits to change?
20:02:17preglowamiconn: you do, but that's just a theory i have
20:02:18goffanot to the proj.ect.. just to his um.. ex wife :)
20:02:27amiconnI am wondering about the h10 disk poweroff...
20:03:16amiconnPrecisely, I am wondering how that works
20:03:21preglowamiconn: where?
20:03:24goffai'd be the perfect candidate.. because my opinion is always the right one :)
20:03:57amiconnlines 77..84
20:05:32preglowamiconn: those ports don't have that
20:05:46preglowamiconn: it seems there are several versions, you see
20:05:55amiconnWhich ports do have that feature?
20:06:04preglowamiconn: if you increment the port address so and so, you get the same port, but with an additional bitmask over the value
20:06:10preglowamiconn: don't remember :/
20:06:25amiconnSounds complicated :/
20:06:43preglowamiconn: i'll see if i can find the code in question, i don't remember where i saw it
20:10:37preglowamiconn: but that h10 code just flips bits the ordinary way, no?
20:11:28amiconnYes it does, but I wonder why that works
20:11:37preglowwhy shouldn't it?
20:11:53amiconnBits 8..15 aren't set that way, so they might be all masked out from changing
20:12:18preglowgpio ports are 8 bit, afaik
20:12:42amiconnYes, but the mask is transferred in bits 8..15 iirc
20:12:54preglowyes, but not in those addresses, i think
20:13:07amiconnSO what do those bits contain when doing an |= or &=
20:13:17preglowzeroes, i'd guess
20:14:08preglowanyway, i think there are at least two versions of those ports, one where you do a regular 8 bit read/write, and one where the write is selectively done based on the mask in bits 8-15
20:14:17preglowbut i can't bloody find the code that made me think so
20:14:28preglowit makes sense anyway, this allows atomic port operations
20:14:48 Quit GFoux (Read error: 110 (Connection timed out))
20:15:07preglowwhich arm can't do otherwise
20:15:27*amiconn wants to find disk poweroff on newer ipods
20:15:28 Quit TCK (Remote closed the connection)
20:15:39preglowyes, nice plan
20:15:40amiconnThe G3 can do that, GPIO_C bit 6
20:15:58 Join GFoux [0] (
20:16:05 Join TCK [0] (
20:16:24amiconn(but currently we don't)
20:16:33amiconndan_a: around?
20:17:02preglowno, i can't for the life of me remember where i saw that
20:17:04amiconnCould you check the GPIO_C value in the debug menu?
20:17:24amiconnThen you know what the state is for hd power enabled
20:17:34amiconn...and could add disk poweroff for the G3
20:17:35 Quit Id2ndR ("Parti")
20:17:49amiconn=> longer battery runtime
20:18:37dan_aI need to get the changes to debug_menu.c to make the ports menu appear committed too...
20:18:39amiconnI'll try GPIO_C bit 6 on my mini g2 as well. If that doesn't work I'll probably open up the thing and the flip port bits and measure
20:18:54amiconns/and the/and then/
20:20:29 Join GreyFoux [0] (
20:21:03 Quit GFoux (Read error: 104 (Connection reset by peer))
20:21:23 Quit _FireFly_ (Remote closed the connection)
20:22:00bluebrothertucoz, back. Had a phone call.
20:22:13tucoz:) hello
20:22:15bluebrotherwhat's exactly your problem?
20:22:37*preglow summons obo
20:22:51tucozwhen I run make on the manual, makeindex complains that it cannot find
20:23:31bluebrotherI assume you have an actual copy? I checked in a small change to LaTeX.Rules a while ago.
20:23:37 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
20:23:45tucozI found a description of that file in a german makeindex manual, and pasted that into a file i named then all is fine
20:23:47 Quit amiconn (Nick collision from services.)
20:23:47 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
20:24:01tucozyes, i found that change
20:24:16tucozbut it looks like it is only me, as the buildserver is doing fine.
20:24:35tucozis that file supposed to be included in a tetex install?
20:25:40bluebrotherno, it's in the manual folder
20:25:53bluebrotherso the makefile should copy it before it starts compiling
20:26:45tucozwonder why i do not have that file
20:27:08tucozis it supposed to be in manual/
20:27:12 Quit My_Sic ("CGI:IRC (EOF)")
20:27:24bluebrotherseems I forgot to check that in :(
20:27:37bluebrotherat least cvs status doesn't know anything about it
20:27:55tucozNo, it is not found when using viewcvs either
20:27:55 Join cucosel [0] (
20:28:28tucozwonder why the buildserver was able to build todays manuals. It breaks when I try to build the manual.
20:28:36 Join bluey- [0] (
20:28:49bluebrotherI knew why I didn't include the index in the cvs builds for now ;-)
20:29:13tucozis that file a style file for the index?
20:29:15bluebrotherhmm, do you have a \printindex in your sources?
20:29:33bluebrotherit controls the output of \printindex
20:29:39tucozhmm. I have tried it, but i have run a cvs up -dPC after that
20:29:51tucozand also make clean in the build-dir
20:30:36amiconntucoz: The build server uses make -k afaik
20:30:42tucozthere. now i got it
20:30:50tucozwhat is make -k?
20:31:31bluebrotherit only breaks on the second run of latex, so there is an output file
20:31:44tucozamiconn, looks like you are right.
20:32:12amiconnIt surely does for the code builds
20:32:42amiconnOtherwise it wouldn't catch all errors a build causes, because it would stop at the first file containing errors
20:32:46 Join spiorf [0] (
20:32:51bluebrotherso the daily manuals may have broken page references
20:32:52 Join akaias2 [0] (n=akaias@
20:32:53tucozaha. so that is the way it works
20:33:04bluebrotherI guess it uses nonstopmode for latex
20:33:19 Join V-t [0] (n=opera@
20:33:31*amiconn thinks that the manual builds should also get a build status table
20:34:11peturor a build table per target (archos/iriver/ipod/x5) ?
20:34:12*bluebrother would really like that
20:34:18 Quit TCK (Remote closed the connection)
20:34:21tucozit would be nice
20:34:23peturit's getting big anyway
20:34:51 Join TCK [0] (
20:35:04peturbut one table is better for the overview
20:37:14 Quit webguest69 ("CGI:IRC (EOF)")
20:37:22 Part V-t
20:37:51bluebrothertucoz, it should work now.
20:37:54 Join pixelma [0] (i=pixelma@
20:37:59tucozyep. thanks
20:38:10bluebrothermy bad, sorry :(
20:39:03tucozwhere do you think we should put the index? after the appendix?
20:39:23tucozI wonder if we should move the licence chapter out of the appendix
20:39:25bluebrotherI put that at the end of the appendix.
20:39:25goffapetur: by build table.. do you mean that you'd have to download the source for each player you wanted to compile?
20:39:38 Quit cucosel ()
20:39:46bluebrotherI was thinking if we should also add the GPL to the licenses
20:39:47tucozto its own seciton
20:39:52tucozyes. i think we should
20:39:53bluebrotheras Rockbox is GPL.
20:40:03 Quit GreyFoux (Read error: 110 (Connection timed out))
20:40:06bluebrotherand I want to modify the license formatting a bit.
20:40:06tucozand we should also add a COPYING file to the manual source tree
20:40:11peturgoffa: no, this:
20:40:16bluebrotherit shows up differently in the tree view.
20:40:37bluebrotherhow about adding some copying-header to every latex file?
20:40:38peturgoffa: and most code is shared between players
20:41:04goffayeah... that's why i asked.. i wasn't clear what you meant
20:41:12goffabut i see what you mean now
20:41:29 Join tim66 [0] (
20:41:59 Join Lear [0] (
20:42:28 Quit akaias2 ("Bye. Need a pastebin?")
20:44:52 Join BigBambi [0] (
20:46:35lowlightI think it would be interesting (though maybe not helpful) to see binary size in the build table, or a better yet a chart of binary size vs time.
20:46:53bluebrothersize vs time?
20:46:59amiconnpreglow: Found the bitmasking stuff. It's the GPIO?_OUTPUT_VAL address +0x800
20:47:06amiconnThe backlight stuff uses that
20:48:25 Quit TeaSeaLancs (Success)
20:49:02amiconnHmm, the +0x800 is true for port B (used for all ipod backlights), but not for port L in backlight-nano_video.c
20:49:07amiconnMaybe this is a bug?
20:49:14tucozlowlight, like x-axis = time, y-axis=size?
20:49:18lowlightbluebrother: to chart the historical growth of the binaries and identify when the increased
20:49:43bluebrotherah, you meant that time −− I was thinking of build time ;-)
20:50:14peturthat would be a great way to confront come committers with their works ;)
20:50:16tucozthen we can calculate a trend, and predict when code size optimizing is needed.
20:51:44lowlightI think code size optimizing is always needed :)
20:52:38tucozwhen the need for code size optimizing is critical then
20:56:13goffadon't lynch me... but would it help things if the voice module were a plugin? I'd think that could clear up a lot of space if people didn't want to include it... (i understand this wouldn't "optimize" things, but would give more room to play i'd think)
20:57:18 Quit TCK (Remote closed the connection)
20:58:13 Join TCK [0] (
20:58:14 Quit _Veseliq_ (Read error: 104 (Connection reset by peer))
20:58:30goffaI'm merely asking... i'm not saying we should/shouldn't
20:59:17peturgoffa: voice is a core feature of rockbox
20:59:54 Join _Veseliq_ [0] (n=veseliq@
21:00:07goffai know it is... but i'm wondering why.. i understand the blind argument, but couldn't we accomplish the same with a plugin?
21:00:08tucozif the voice were loaded as a tsr-plugin, wouldn't that stop other plugins from working?
21:00:32goffaor would that have undesired side effects?
21:00:44goffaother than just the concept of removing voice
21:01:11 Part Genre9mp3
21:01:13amiconnYou can't move voice out of the core
21:01:21amiconnIt's tied to it
21:01:48 Join Genre9mp3 [0] (
21:01:49amiconnThe menus access it, it needs the codec infrastructure on swcodec etc
21:02:28goffaok... then i have to ask... why was it designed that way? (not being critical.. just trying to understand)
21:02:41amiconnbecause it has to
21:04:25amiconnFor the menus it might not be that important, but the voice playback itself needs to be coordinated with the rest of the audio playback
21:04:59goffaah.. so in order for voice to function.. it has to be included... but for the player to function voice doesn't need to be included?
21:05:44amiconnNo, the voice code could be cut out completely, leaving a voice-disabled rockbox
21:06:01goffaok.. that's what i meant with the last statement.. you worded it better
21:06:13goffabut that's what i was asking
21:06:29amiconnBut I consider voice a key feature, more important than many others
21:06:42goffai understand your feeling on that
21:07:10goffaI was just trying to get a grasp on why it had to be part of the i think i understand
21:07:40goffawould it be a major pain in the ass to have a compile time option to disable it if it wasn't desired?
21:08:59goffaor maybe one exists and i'm not aware of it (that could be entirely possible)
21:09:54amiconnThere is no such option
21:11:06goffaI guess i'm indifferent to it, but unless i shoot my eye out, i can't see using it :)
21:12:51lowlightgoffa: Since the voice files are loaded from disk and use the audio codec & playback engine, I don't think there's that much overhead, especially compared to some other options.
21:12:52*amiconn uses voice in the car
21:12:57 Join Paul_the_Nerd [0] (
21:13:45goffayeah... i'm starting to understand.. i just wasn't sure how things were set up
21:14:18LearJust don't install any voice file and don't worry about it.
21:15:09 Quit Rob2222 (Read error: 104 (Connection reset by peer))
21:15:12goffaLear: nah... not good to remove it if it will break the functionality altogether
21:15:38goffaunless you mean delete the wavs
21:15:48goffabut i don't think that would gain you much
21:15:52 Quit charkins ("Ex-Chat")
21:16:14 Join Rob2222 [0] (
21:16:17LearI mean the voice file(s) in ./rockbox/lang (or whereever).
21:16:27goffaah .. ok
21:16:39LearOn software codec platforms, if that a voice file is found, memory is reserved for it, iirc.
21:16:43goffalike with the fonts
21:16:53 Join charkins [0] (
21:17:07 Join Arathis [0] (
21:19:08 Quit charkins (Client Quit)
21:19:24preglowamiconn: where'd you find it?
21:19:28preglowamiconn: also, did you check how it works?
21:21:40 Quit rretzbach (Read error: 104 (Connection reset by peer))
21:31:55preglowahhhh, yes
21:31:57preglowthis is where i saw it
21:32:30preglowyou think we should toss in some new defines for this?
21:32:55 Quit TCK (Read error: 110 (Connection timed out))
21:33:38dan_aThat line doesn't make a lot of sense to me - why does it do that instead of "GPIOB_OUTPUT_VAL |= (1 << 3)"?
21:34:07 Join MadDog011 [0] (
21:34:15 Part MadDog011
21:34:37amiconnI would like to see the GPIO L behaviour on nano checked
21:34:42preglowdan_a: because the line you propose turns into a read, an or then a write
21:34:48preglowdan_a: when you instead can do just one write
21:34:58preglowamiconn: what behaviour?
21:35:14dan_aAh. Thanks, preglow.
21:35:15preglowdan_a: a read, an and, then a write
21:35:21preglowi mean or, barrghg
21:35:22preglowi suck
21:35:27preglowyou get what i mean :)
21:35:34amiconnThe fact that the GPIOL_OUTPUT_VAL address isn't midified by the 0x800 offset
21:35:49preglowdan_a: this way, the port manipulation is atomic. so that nothing can happen between the read and the write
21:35:57amiconnIiuc this should write the (8-bit)value directly instead of masking
21:36:29preglowamiconn: where is it used?
21:36:48amiconn[20:48:45] <amiconn> Hmm, the +0x800 is true for port B (used for all ipod backlights), but not for port L in backlight-nano_video.c
21:36:57preglowright, missed that
21:37:10 Join mirak [0] (
21:37:44 Quit bluey- ("Leaving")
21:38:43preglowamiconn: perhaps the pp chip checks if bits 8-15 are all 0, and then just drops the masking. a value of 0 there would make no sense anyway
21:39:06amiconnWhy the offset for port B then?
21:39:07preglowaliased addresses is surprise in pp
21:39:28preglowprobably just an ordinary decoder logic glitch/don't care case
21:39:55preglowthis should be easy to check
21:39:56amiconnRiight, dontcare bits are rather common it seems
21:40:19preglowthat is, if i find a port that isn't important :>
21:42:53***Saving seen data "./dancer.seen"
21:48:31 Join edx [0] (
21:49:52preglowshouldn't the "deleted xxxxxxxx" screen saying "deleting xxxxxx" ?
21:50:08preglowit's really annoying when deleting big directories, it seems like rockbox has hanged
21:50:17Paul_the_NerdIt really should say deleting
21:50:37tucozi had forgot about that, but that should be fixed
21:51:07preglowanyone keen on doing so? i'm allergic to lang files
21:52:08amiconnThe delete screen should show progress of some sort
21:52:32amiconnWe don't know the file/dir count in advance, but it could show the raw count as it goes
21:52:45 Quit lightyear (Remote closed the connection)
21:53:02 Join rconan [0] (
21:53:11amiconnOf course it would be better if it would be faster though. Dunno what causes rockbox' recursive delete to be that slow
21:53:15 Join Rincewind [0] (
21:54:09 Quit Kitt0s ("( :: NoNameScript 4.02 :: )")
21:55:39preglowyeah, i was wondering about that
21:57:18preglowprogress really isn't so important, but a way of showing that rockbox is alive is truly needed
21:57:27preglowthe same problem applies to copy/pasting big directories
21:57:29preglowwhere it's even slower
21:57:48preglowi think there's even a patch in the tracker that adds messages to those operations, as those currently don't display anything
21:57:52preglowwhich is very bad
21:58:31 Join |AhIoRoS| [0] (n=ahioros@
22:00:27 Join jhMikeS [0] (
22:01:39amiconnjhMikeS: What was the problem with the asm i2c code?
22:01:42*amiconn is curious
22:02:34jhMikeSThings like: btst.l #12, %%d0 <= wrong bit
22:03:29preglowamiconn: remove_dir is a very short function, so some of the basic file/dir io operations in rockbox have to be very slow
22:03:30jhMikeSSo now it's: btst.l %[sclbnum], %%d0 with: [sclbnum] "i"(SCL_BITNUM)
22:05:05 Quit qailer ("Leaving")
22:05:38preglowamiconn: why are the backlight-*.c functions marked inline? they're alone in a .c file...
22:06:22amiconnUmm, ask barrywardell. He did the target tree transition for ipod
22:06:22*jhMikeS still can't believe that one waited so long
22:07:12jhMikeSNow full screen update fr is 61.5 fps unboosted. Faster than the refresh rate.
22:07:18 Quit bluebrother ("Leaving")
22:07:23preglowhaving them inline is good, but then they should be in the h file
22:09:21 Join barrywardell [0] (n=barrywar@
22:15:21 Part Rincewind
22:21:10preglowamiconn: changed both the values for nano around +/- 0x800, and it had nothing to say
22:21:22preglowso things speak in favour of a don't care bit here
22:21:55 Join webguest50 [0] (i=ddba59a3@gateway/web/cgi-irc/
22:23:17preglowseems like there are several...
22:23:29webguest50just thought I'd pop in and say thanks for the crossfeed feature, it works very well on old recordings, Mamas and the Papas sound good
22:23:45preglowwebguest50: you're welcome, it does indeed :)
22:24:07webguest50wizadry indeed :-)
22:24:56preglowwhat player've you got?
22:25:08preglowahh, it works nice and fast there, yes
22:25:16webguest50err two of 'em
22:25:31preglowas it should be
22:25:39 Join RoC_MM [0] (
22:25:40preglowi wish i had two
22:26:08webguest50the other one is just incase...well just for incase
22:26:08 Quit RoC_MM (Client Quit)
22:26:24preglowi don't think i'll find another dap that's nicer any time soon
22:26:53webguest50yes a class player for sure
22:27:24preglowamiconn: an amazing amount of don't care bits...
22:27:46 Quit akaias ("Bye. Need a pastebin?")
22:27:57preglowamiconn: address + 0x200, 0x400, 0x800, even 0xe00 works fine.....
22:28:12preglowwebguest50: it's also got more than enough cpu for even fancier stuff than crossfeed
22:28:47webguest50well I dont really use crossfeed, but do use Replay Gain and Eq constantly
22:28:59webguest50err thats crossfade
22:29:17preglowjlo actually wanted to make the crossfeed a bit fancier than it is, but i had to nerf it a bit to make it work fast enough on stuff like ipods
22:29:41Paul_the_NerdYou think you'll unnerf it after COP support maybe?
22:29:46preglowno :-)
22:30:02preglowhe wanted fancier filters, which i don't really think does much for the sound
22:30:06preglowi think he eventually said so as well
22:30:28Paul_the_NerdWell that's fine then
22:30:31 Join PaulJam [0] (
22:30:54preglowi'll probably do some eq work next
22:31:11 Join akaias [0] (n=akaias@
22:31:11barrywardellpreglow: that backlight code is still inline because I didn't change it when moving it to separate files
22:31:28barrywardellit wasn't something i decided on, it's just something i didn't notice
22:31:32preglowamiconn: well, then i most certainly learned something new today
22:31:39webguest50Rockbox is just darn good, nice features and great sound
22:31:40 Join midgey34 [0] (
22:32:03preglowoh, this reminds me dithering is still disabled :-)
22:32:10preglowwe really should make that a new option
22:32:27barrywardellamiconn: when I was looking for the H10 ata power off bit it was helpful to read all the GPIO_OUTPUT_EN to see which bits had been enabled for output by the bootloader
22:32:39barrywardellthat narrowed things down a lot
22:33:44 Join Kitt0s [0] (
22:34:34 Join matsl [0] (
22:34:45preglowhaha, i don't think dithering has been enabled since i disabled it for being too slow with mp3 a long, long time ago
22:36:05preglowsomething i can guarantee you it's not
22:36:21 Quit akaias ("Bye. Need a pastebin?")
22:36:48preglowbarrywardell: well, might as well remove the 'inline' if they're not supposed to be inline
22:37:09 Join Id2ndR [0] (
22:37:15webguest50thanks. later
22:37:18 Part webguest50
22:37:51barrywardellpreglow: they were inline back before moving to target tree. i just copied them directly into the separate files
22:38:11barrywardelli don't know how much benefit there is to inlining them
22:38:11 Join nudelyn [0] (
22:38:27preglowdepends on the platform
22:38:33preglowno, not really
22:38:50preglowthey are just used one place, i think, so inlining might be a small net gain
22:39:06preglowno, they're not...
22:39:19preglowlet's just keep them where they are
22:39:39Learpreglow: just moving the dither structs to iram made a big difference.
22:40:11barrywardellpreglow: feel free to change them if you want if you feel it will make a difference
22:40:19preglowLear: all dsp should be coded in asm anyway
22:40:38preglowbarrywardell: no difference, no, it's purely syntactic nitpicking
22:41:01preglowi've got some dsp.c restructuring coming up one of these days
22:41:32Learyou've been talking about that for a while... :)
22:41:46preglowyeah, but i haven't had time for any rockboxing for a long while now
22:42:16preglowouch, dither_sample is called per sample
22:42:21preglowthat's asking for slowness
22:46:13preglowwe do second order noise shaping too, apparently
22:49:21 Join akaias [0] (n=akaias@
22:49:40preglowbut yeah, everyone do agree with me that allowing each decoder to decide whether dithering is needed or not is foolish, yes? it should be a user option?
22:52:09 Quit Arathis ("[rl_bot quit]")
22:52:20 Quit jhMikeS (Read error: 104 (Connection reset by peer))
22:52:36Learmakes sense. possibly if codecs could be allowed to disable it, for "known slow" cases or something?
22:53:11preglowi really don't think so
22:53:30preglowyou need to disable all other stuff that make something too slow yourself
22:53:30 Quit Quarryman ("bah")
22:53:41preglowso dithering shouldn't behave any differently, for consistency
22:53:44barrywardellpreglow, amiconn: what's this about masked writing to gpio?
22:54:04preglowbarrywardell: if bits 8-15 are non-zero, they decide which of the eight lower bits actually get written
22:54:44preglowbarrywardell: so if you write 0x101 to a gpio reg, the lowest bit of the reg gets set
22:54:49preglowand the other bits aren't affected
22:55:02 Join jhMikeS [0] (
22:55:08barrywardellah, i see. sounds useful
22:55:12preglowbarrywardell: indeed
22:55:17barrywardelllearn something new every day :)
22:55:20preglowbarrywardell: just a theory so far, but everything i've seen says it's correct
22:55:35barrywardelli'll test it a bit and see if i can veryify
22:55:46barrywardellshould be fairly easy to test
22:55:53 Part lowlight
22:56:01 Quit rconan (Read error: 104 (Connection reset by peer))
22:56:55preglowbarrywardell: it also seems that many port addresses have a very large amount of don't care bits
22:57:15barrywardellwhat do you mean by that?
22:57:38preglowbarrywardell: for instace, writing to GPIOB_OUTPUT_VAL does the same even if you add 0x200, 0x400 or 0x800 to the address of it before writing
22:57:50preglowbarrywardell: so there's a big amount of aliased addresses, it seems
22:59:04barrywardellah right, i see
22:59:23barrywardelli wonder if they're used for different purposes?
22:59:56 Join arkascha [0] (
23:00:13preglowi just think it's for simplifiying the decoder logic in the pp
23:00:28preglowyou save a couple of gates for each bit you just don't care about in the address
23:00:51barrywardellah, ok
23:01:20 Quit bundik ("Quit")
23:02:48 Quit Lear ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
23:07:52 Quit barrywardell ()
23:08:45 Join RogerBacon [0] (
23:10:26 Quit matsl (Read error: 110 (Connection timed out))
23:11:06 Join matsl [0] (
23:13:52 Quit lee-qid ("Trillian (")
23:17:48 Quit arkascha (Remote closed the connection)
23:21:16 Join NickDe [0] (
23:21:42 Join pagefault [0] (i=pagefaul@pdpc/supporter/active/pagefault)
23:23:21midgey34would a linux user be able to test a patch for me?
23:23:39midgey34it fixes the sim building on osx but it may cause problems elsewhere
23:23:54midgey34it shouldn't, but i'd like to make sure
23:24:24Paul_the_NerdYou could always just try it in the Debian VMWare image, couldn't you? ;)
23:25:33midgey34heh, i could but it will have to wait for another day
23:25:40midgey34exam tonight
23:27:16 Quit akaias (Read error: 104 (Connection reset by peer))
23:27:49 Join akaias [0] (n=akaias@
23:30:10 Quit petur ("Zzzz")
23:33:55BigMacPaul_the_Nerd:I know I am not one to talk, but why do you talk like Will shatner in star trek when your on the forums
23:34:06BigMaclike puting ** on random words
23:34:08 Quit midkay (Read error: 104 (Connection reset by peer))
23:34:09 Part RogerBacon ("Quitte")
23:35:04*Paul_the_Nerd shrugs
23:35:31Paul_the_NerdIt's a habit I picked up.
23:35:32BigMacjust wondering if you had some logic behind it
23:35:37BigMacOh alright
23:35:48Paul_the_NerdWhen I want to make sure someone is clear that something is either important or emphasized, I use that instead of bold.
23:35:55Bagderthat's a gimmick all undercover FBI a gents use
23:36:08Paul_the_NerdThey're not supposed to know!
23:36:23amiconn*something* actually means bold in some plaintext email clients
23:36:27BigMacI could see if it was important words in the sentence
23:36:27Bagderdarn, luckily I slipped in a space so that the autodetection didn't trigger
23:36:32BigMacbut you write like
23:36:35blue_lizarduuuuu FBI uses rockbox
23:36:37Paul_the_Nerdamiconn: That's kinda where I picked it up.
23:36:44Paul_the_NerdPlaintext various things.
23:36:46blue_lizarddidnt knew the are intelligent at all
23:36:54BigMacI once went to the carnival *that* had the best cotton candy
23:37:07amiconnSame way there is _underline_
23:37:12Paul_the_NerdUnderline == italics
23:37:18BigMacI usually read your message with the Will shatner voice
23:37:23Bagdergood old BBS traditions
23:38:01 Join matsl_ [0] (
23:38:16hcsBe *careful* where you *place* emphasis.
23:38:57Paul_the_NerdHmm, of my last 15 posts, only two use *s for emphasis, and I think they're on good words.
23:39:24preglowhmm, third order noise shaper
23:39:26 Quit matsl (Read error: 110 (Connection timed out))
23:39:43 Quit secleinteer (Read error: 104 (Connection reset by peer))
23:39:45*preglow misses bbses
23:39:53Paul_the_NerdWhat do second / third order mean?
23:40:02BigMacoh btw
23:40:10BigMacThank you Paul_the_Nerd for the battery tips
23:40:20BigMactoday I spent 2 hours at the gym
23:40:27BigMacleft with 74% battery
23:40:34*Febs remembers dialing in to BBSes with a 100 baud modem.
23:40:43preglowPaul_the_Nerd: essentially it just describes how sophisticated the noise shaping filter is
23:41:04Paul_the_Nerdpreglow: Ah, gotcha.
23:41:17preglowlowest modem i can remember using is 2400 baud
23:41:21preglowlowest speed, that is
23:41:32Paul_the_NerdEither 2400 or 1200 here, I can't really remember which.
23:41:33BigMacI think it was the tag cache that was killing me
23:41:34amiconnPaul_the_Nerd: YAM uses *bold* /italic/ and _underline_
23:41:39FebsNow I feel old.
23:41:43Paul_the_NerdBut I was overseas, so it was more availability than state-of-technology that restricted me
23:42:19Paul_the_Nerdamiconn: _ for underline makes more sense than for italics anyway.
23:42:55***Saving seen data "./dancer.seen"
23:43:37z0deis there a play random album(folder) option in rockbox.
23:43:46Paul_the_NerdThere's a patch for it.
23:43:53PaulJamz0de: yes
23:44:02preglowwas leet anyway
23:44:04PaulJamPaul_the_Nerd: it is in cvs
23:44:08preglowthere i picked up my penchant for ansi graphics
23:44:08Paul_the_NerdIt is?
23:44:11Paul_the_NerdI thought it got rejected
23:44:19Paul_the_NerdDid it get un-rejected?
23:44:20z0decvs=bleading edge?
23:44:27preglowPaul_the_Nerd: it just got commited
23:45:58tucozFebs, 100baud. When was this?
23:46:05PaulJamz0de: you can use a bleding edge or daily build.
23:46:19z0defor random folder play
23:46:35z0dei cant find it in the build i have (yesterdays dally build)
23:46:46Bagderthey were 110 baud iirc
23:46:51PaulJamz0de: before you can use the feature you must run the random_folder_advance_config plugin
23:47:49z0deill do that, thanks PaulJam
23:48:03PaulJamz0de: the option is under general settings => playback => autochange directory or similar
23:48:53z0debut thats after the plugin?
23:49:02 Part tim66
23:49:41*tucoz understands that baud is a more complex term than bps
23:49:42PaulJamyou can enable it before, but it wont work until you run the plugin
23:50:00preglowtucoz: baud is symbols per second, bps is bits per second
23:50:19Bagderback in the early days, you got a bps per baud
23:50:25tucozpreglow, i know that now. i am reading the wikipedia articel
23:50:28preglowthat's really long ago, heh
23:50:34preglowin the days of fsk
23:50:44tucozso one symbol == 1 bit?
23:50:49preglowback then, yes
23:50:57preglownow, not by a far shot
23:51:10tucozwhat about the 9.6, 28.8 etc modems?
23:51:19tucozwas that data rate, or was it baud?
23:51:24preglowdepends on what they do, but they're usually multibit psk coders
23:51:41tucoz^^to bagder
23:51:43preglowwhich map several bits data to each signal transition
23:51:52z0dehow do i get the plugin i've found the info in the wiki but no plugin
23:52:07tucozz0de, what do you mean?
23:52:30z0dewhere do i get the plugins
23:52:35tucozpreglow, aha. ok.
23:52:41 Quit random81 ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
23:52:46tucozz0de, those are included when you install rockbox
23:53:46tucozas far as i know, there are very few unofficial plugins available.
23:53:58tucozand the official ones are included in the
23:54:05Bagderand those that do exist run out of date in no time
23:54:40tucozyes. that is why we should release rockbox sometime :)
23:54:43z0deso its all on the player already
23:55:04BagderRockbox is our prisoner, let's never release it
23:55:07tucozwhat plugin are you thinking about, and what player do you own?
23:55:22z0dex5, random album playing
23:55:36amiconnWow, now I am impressed. Throwing out all '@echo..' from apps/plugins/Makefile shortens plugin build time by ~16%
23:55:52Bagdercygwin crazyness
23:55:52amiconn...without changing the output
23:56:10tucozz0de, that should be included for all targets
23:56:10amiconnI think it helps on linux as well
23:56:19Bagderamiconn: yes, but hardly 16%
23:56:21preglowamiconn: doubt it
23:56:36z0deyea ive found it
23:57:16amiconnI just tested with apps/plugins. As it works rather well, I think I should apply it to the other Makefiles
23:57:24 Quit Soap (Read error: 110 (Connection timed out))
23:58:23preglowhmm, should noise shaping be a separate option from dithering?
23:58:25tucozamiconn, do you show info some other way, or do you run really silent?
23:58:32preglowi'm kinda leaning towards having noise shaping always enabled
23:58:34amiconn[23:55:08] <amiconn> ...without changing the output
23:58:50tucozmissed that

Previous day | Next day