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 2007-03-08

00:01:27LinusNDataGhost: reading 2 sectors from sector 63 looks pretty unaligned to me
00:01:35DataGhostthat's intentional
00:01:51DataGhostI don't know if you know but the 80GB plainly refuses to read anything
00:01:52DataGhostoh wait
00:02:05DataGhost63 hm? that's probably your partition guessing code :)
00:02:11DataGhostit should find it at 252
00:02:40jhMikeSdan_a: well, both database and recording get stuck if you record during database update. The disk writes just stick and the drive twitches about once per second after that. Dpm
00:02:42 Quit OgMaciel ("mv OgMaciel $HOME")
00:04:25LinusNtime to get some sleep
00:04:32LinusNnite all
00:04:32 Part LinusN
00:04:32 Join jhulst [0] (
00:04:33jhMikeSDon't know if this affects anything other than Coldfire. Memory conflicts don't seem the reason though.
00:07:20 Join Rondom [0] (
00:08:28jhMikeSBut afaik it's the heaviest case of two threads writing to disk at the same time.
00:10:05 Quit Farp ()
00:11:17 Quit lee-qid ("aufwiederbyebientotsayonara")
00:16:46 Join iainneedshelp [0] (i=3e1ff264@gateway/web/cgi-irc/
00:17:30 Quit iainneedshelp (Client Quit)
00:17:32 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
00:17:35 Join iainneedshelp [0] (i=3e1ff264@gateway/web/cgi-irc/
00:18:09iainneedshelpHey all can anyone help me? I have a question!
00:18:33 Quit linuxstb ("CGI:IRC (Ping timeout)")
00:18:57 Quit iainneedshelp (Client Quit)
00:19:00 Join iainneedshelp [0] (i=3e1ff264@gateway/web/cgi-irc/
00:19:09linuxstb_iainneedshelp: Just ask your question.
00:19:10 Quit iainneedshelp (Client Quit)
00:19:18safetydanHrmm... I thought the speex decoding stereo issue was fixed? Darn.
00:19:55 Join muesli__ [0] (n=muesli_t@
00:20:58*dan_a bed
00:21:13safetydanDoes pcmbuf_insert expect length in bytes or samples?
00:21:23jhMikeSper channel
00:21:49safetydanI think that might be why speex stereo doesn't work then.
00:24:12jhMikeSit's sending twice the count for stereo? (guess)
00:25:03safetydanI think it's actually half what it should be for stereo
00:25:08safetydansee line 555 of speex.c
00:27:08*preglow cranks up his asm brain centre
00:27:31 Quit dpmarsh ("nite nite")
00:28:12preglowin a related note, i found the amount of smull and smlal in retailos to be very small
00:28:31preglowalmost gives me a small hope pp has hidden away some fast mac stuff somewhere
00:29:08 Quit Rondom ("Ex-Chat")
00:29:20preglowi found a couple of exact equivalents of our eq_filter and some long streaks of what seemed to be fir filtering, but not much more
00:29:30 Quit muesli- (Read error: 110 (Connection timed out))
00:29:31jhMikeSframe_size should be the same for stereo and mono
00:29:42jhMikeSI can't tell if it is though
00:30:08jhMikeSpreglow: which retailos? ipod?
00:30:13preglowjhMikeS: aye
00:31:32jhMikeSsafetydan: or should say new_frame_size should be the same for both
00:32:40jhMikeSpreglow: it's not like we should expect them to be as concerned about optimizing as we
00:32:54safetydanjhMikeS, actually now I'm not sure what's going on. Can't really look at very hard as I'm at work and not near a C compiler :)
00:34:09 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
00:34:26preglowjhMikeS: still doesn't really explain why there's so little smlal and smull is only some speex files? I'm just thinking you might have to /2 for stereo if libspeex gives samples*channels
00:35:12jhMikeSpreglow: they do everything in 8 bits? :p
00:35:46 Quit Nico_P (Remote closed the connection)
00:35:57preglowjhMikeS: heh
00:36:26jhMikeSit litterally takes hours for database to update in the bg on my x5 with 40GB of music
00:37:33preglowit doesn't even use five mins here with 20 gigs
00:37:47preglowthat's idling, though
00:37:57jhMikeSit's a background update
00:38:13 Quit |Rincewind| ("Cya")
00:38:16preglowi don't really know how useful i think a background update is, though
00:39:08 Quit inversions (Read error: 110 (Connection timed out))
00:39:10 Quit muesli__ ("ich will Kühe!!!")
00:39:18 Join inversions [0] (
00:41:08jhMikeSit causes the iRivers to hang if you record during it (it shouldn't do that). It just causes a Encoder buffer overflow warning on my x5 but no hang ... so I'm happy with that. You shouldn't really record during that anyway.
00:41:41preglowi think we should make it a foreground process anyway
00:41:46 Quit ompaul ("enjoy")
00:41:59preglowif only to eliminate chances of stuff like that happening
00:42:02DataGhostgrmbl... is there some generic function already in the rockbox code to display debugging output on the screen?
00:42:04jhMikeSRecording really needs your disk to be quiet
00:42:27 Quit Arathis ("Bye, bye")
00:42:30safetydanDataGhost, logf?
00:42:34DataGhostbecause the one I had to write now (normal debugging code doesn't look as if it's intended for use without a serial port / debugging program)
00:42:35preglowDataGhost: well, if you want it on the screen, you probably need to do snprintf, lcd_puts, lcd_update
00:42:38jhMikeSDataGhost: not on the main lcd but you can use splash screens to help\
00:42:38DataGhostis terribly slow
00:42:39preglowDataGhost: but logf is nice
00:42:52DataGhostwell I'm using the snprintf puts and update thing now
00:42:58DataGhostbut it doesn't work well when put in a function
00:43:05preglowsplash is nice
00:43:11DataGhostwhen I put the code directly where I need it, it's almost so fast that I can't see it
00:43:19DataGhostand now it 'scrolls' at 1 line per 10 secs
00:43:22preglowlogf with a remote is optimal, but you're on ipod, so no dice there
00:43:38jhMikeSdead loop while(++x < num); so you don't have yields where you don't want em
00:44:16DataGhostwell I actually also meant to say that I can write only to a specific line on the screen and that I want it to 'scroll'
00:44:31DataGhostso it's actually pretty hard when writing this from 2 different functions
00:44:37 Join TrueJournals [0] (
00:44:40DataGhostso I have to divide the screen in two
00:44:46jhMikeSI'm guessing something to reserve a portion of the display for debug output is in order
00:45:07DataGhostoh i'm just using the entire display now
00:45:36DataGhost <- that's what I want... works great but it's only called from 1 function.. it's even just coded into that function
00:45:36jhMikeSA logf build wihtout a remote ought to do that imho
00:45:47 Join dmdfan [0] (
00:45:52TrueJournalshey... I have a quick question about FS #6718...
00:46:05DataGhostbut I'll try that then
00:46:32TrueJournalsWhy is being able to re-order menus frowned upon by "core developers"?
00:46:48 Quit roolku ()
00:46:53preglowTrueJournals: support nightmare
00:47:00 Quit mirak (Read error: 104 (Connection reset by peer))
00:47:06TrueJournalsdarn stupid people... :-p
00:47:10TrueJournalsI understand though
00:47:18preglowwe try not to make our lives too difficult for ourselves
00:47:41TrueJournalsOK, I guess that makes sense
00:47:49preglowand we also feel saving a couple of button clicks here and there is worth it
00:47:50nlsit's pretty non-KISS too...
00:48:05preglowis NOT worth it
00:48:17jhMikeSI'm sure menu reordering could be implemented very simply
00:48:27linuxstb_I'm not saying I think it's a useful feature, but why would re-ordering the root menu be a support problem? We never tell people to click on item N in a menu, we give the name...
00:49:00 Join Thundercloud [0] (n=thunderc@
00:49:03DataGhostso jhMikeS a logf build outputs debugging information to the screen by default, or what?
00:49:15safetydanlinuxstb, apart from blind users, though I doubt they'd be customising build order
00:49:19jhMikeSDataGhost: I don't believe so unless you have a remote
00:49:21safetydanmenu order
00:49:41DataGhostgah. I thought you said it should be able to do that when I don't have a remote
00:49:53bluebrotherusers will want to remove unused items from "their" menu
00:49:56jhMikeSI'm saying it probably should be a feature that gets added though
00:50:06DataGhostah :)
00:50:07linuxstb_bluebrother: So? We don't let them do that.
00:50:10bluebrotherif that is not implemented they will scream again and again
00:50:25DataGhostwell at least now that I know that logf should do that, maybe I can move that debugging-to-the-screen over there
00:50:27bluebrotherif we don't let them do that, why do they need reordering of the menu itself?
00:50:49linuxstb_But the argument is that it's a support nightmare...
00:50:55bluebrotherMS word is capable of completely reorganizing the menus. How much users actually do that?
00:51:00linuxstb_I'm just saying I disagree with that reason.
00:51:44bluebrotherthe names might be not a good reason for rejecting reordering of the menus
00:51:48TrueJournalsbluebrother: I really want to re-order the menu just because it'd be more convienient for me to have Database first in order because I always use that, and rarely use the files browser
00:52:00nlsPersonally I don't think it is a bad feature, I just don't like it ;-)
00:52:10bluebrotherTrueJournals: still I don't see your problem. Just set the database as startup screen
00:52:21jhMikeSDataGhost: you probably could add a logf overlay to the display pretty quickly...use a separate buffer for it, and update it in lcd_update_rect and lcd_update before the driver writes to the device.
00:52:22TrueJournalsIt's just annoying to always have to hit down in order to get to my music... especially because the H10 touchpad is so sporadic
00:52:28bluebrotherand when the menu comes in the last used item is selected, which is DB
00:52:37bluebrotherwhich is right next to "now playing"
00:52:48preglowlinuxstb_: afaik, i'm the only one who said that, and i was just parroting what i thought i remembered someone else saying
00:53:01linuxstb_preglow: bluebrother said it as well on FS.
00:53:02preglowmy memory, however, should never be trusted
00:53:05preglowah, ok
00:53:33TrueJournalsbluebrother: but I don't always want to go to the databse... I want to see the root menu... it's nice to have, but I would like it to be set up how I want it, not how someone else wants it...
00:54:00linuxstb_TrueJournals: It's very easy to change if you compile Rockbox yourself...
00:54:02bluebrotherwhat's against having a menu in a way that makes it most versatile?
00:54:03safetydanTrueJournals, the change you're talking about should be easy enough to make just in your build of Rockbox.
00:54:39TrueJournalsWell, I gotta go... sorry
00:54:41TrueJournalssee ya
00:54:41bluebrotherI'm completely against such a feature in the core. If someone wants to make a build of his own, fine.
00:54:49 Part TrueJournals
00:56:28bluebrotheranyway, time for sleep. bye
00:56:31 Quit bluebrother ("leaving")
00:57:48safetydangrrr, sometime I'm going to remember to fix playlist catalog not to bail straight back to the WPS when you press left
00:57:56jhMikeShrm...x5 has no locks difficulties whatsoever... but background database updates have to defer to more important activity
00:57:58*DataGhost bows to jhMikeS
00:58:41jhMikeSThen there's making the display smaller changing LCD_WIDTH/LCD_HEIGHT
00:58:55DataGhostI just copied the logf code responsible for writing to the remote
00:59:00DataGhostadjusted the settings and function names
00:59:01DataGhostbrilliant :)
00:59:22DataGhostnow I just need to make DEBUGF point to logf
00:59:23jhMikeSso it's working for you?
00:59:26DataGhostyeah :)
01:00:06*petur copies music to his h380 :D
01:00:59jhMikeSpetur: so you should have that filled in a couple days?
01:01:09peturhope not ;)
01:01:42jhMikeSpetur: I don't think there's any recording concern unless you have database updating in the background no worry
01:01:59peturI have database disabled
01:01:59jhMikeSbut that's just nuts to do anyway
01:02:23DataGhostthere's still a minor buffer-bug in there (I guess)
01:02:27DataGhostbut at least it works :)
01:03:22jhMikeSIt limits logf to 30 chars wide and I don't know if it assumes certain display size somewhere
01:03:51DataGhostit does
01:03:52linuxstb_petur: What's the current price of an 80GB disk? Did you buy your disk new?
01:03:52 Part nls
01:03:54DataGhostwell, mine does
01:04:11DataGhostbut logf is limited too.. ok I'll look into that :)
01:04:27peturlinuxstb_: it was new, bought it on ebay for $260
01:04:55peturthere were 3 on sale and all are now gone
01:05:28DataGhostwell, that was easy
01:08:06 Join JdGordon [0] (i=82c20d65@gateway/web/cgi-irc/
01:08:13JdGordonhey all
01:08:57***Saving seen data "./dancer.seen"
01:09:21DataGhost :) now that's all sorted out, maybe I can make some progress
01:10:04jhMikeSthat looks nicer. I suggest the overlay so you can move it or hide and see the UI at the same time.
01:10:18DataGhostwhen I see the UI, it's too late
01:10:31linuxstb_petur: Is it the mk8009gah ?
01:10:48DataGhostit's for the 80GB... so I need to see all pre-gui FS init stuff
01:10:57peturMK8007GAH (the later ones have the wrong connector)
01:11:08peturfor my h3x0 at least
01:11:19 Join straylight42 [0] (
01:11:42linuxstb_OK. Because there are a lot of mk8009gah drives for USD $169.99 on ebay at the moment...
01:12:14peturwon't work in h140/h340
01:12:15linuxstb_The listing says it works on a video ipod.
01:12:24 Quit Thundercloud (Read error: 104 (Connection reset by peer))
01:13:00 Join Thundercloud [0] (n=thunderc@
01:14:46peturlinuxstb_: I got one from this company:
01:14:57*petur discovers price went down :(
01:16:25linuxstb_petur: They were the people selling the mk8009gah as well...
01:16:26 Quit Thundercloud_ (Read error: 110 (Connection timed out))
01:16:50preglowwhat's the largest disk i can stuff in my h120?
01:16:56preglow30 gigs?
01:17:21peturl don't know if the 40gig single platter is available already
01:17:49 Join Shaid [0] (
01:17:57*linuxstb_ is annoyed that 2.5" PATA drives seem to have stopped at 160GB
01:19:18jhMikeSI think database updates have to be pauseable/resumable and something is borked with ata throughput when more than one thread is writing alot. :\
01:19:18preglowi should just stop messing around and start looking for an h140
01:19:24preglowthis thing ain't gonna live for ever
01:19:35preglowjhMikeS: seek overhead?
01:20:09jhMikeSnot sure doesn't sound like the head itself is getting much workout
01:20:20peturlinuxstb_: mk8009gah and MK8007GAH have a different connector, so it depends on what player you need it for
01:20:48DataGhost That's what I was looking for jhMikeS :P
01:21:17DataGhostnow I think I'll go to bed because my camera needs to charge
01:21:32peturhahahaha 'This is not FAT32. Go away!'
01:22:12jhMikeSit shouldn't lock the iRivers so bad at the least though. if recording can pause the activity that would be the right thing anyhow.
01:22:41linuxstb_petur: I'm undecided atm - it could be nice in a gigabeat...
01:23:05jhMikeSif it were just seek overhead then playback should have trouble but it doesn't
01:23:06Shaidgigabeats take 2.5" drives?
01:23:19preglowthat would surprise me if so
01:23:39linuxstb_Shaid: No.
01:23:39peturdoesn't the gigabeat have a 50pin ata connector? then the mk8009 is useless...
01:24:20 Join [Tesser] [0] (n=Tesser@unaffiliated/tesser/x-000001)
01:24:29peturand can you format a new drive from the gigabeat or over usb?
01:24:38 Join p0ser [0] (
01:24:39JdGordonguys FS6757, pre root menu, starting without headphones plugged in would put you in the file browser if this setting was set.. what should it do now? just put you in the root menu?
01:25:15jhMikeSpetur: I think so, yes
01:25:42linuxstb_JdGordon: I would say it should go in the startup screen - that's what the file browser used to be.
01:26:10JdGordonright, but in this case the start screen is set to wps...
01:26:25linuxstb_Then yes, the root menu makes sense.
01:27:33toffe82petur : if you put a new drive in a gigabeat , you can do nothing, you need a special progam from toshiba to make it bootable
01:28:06*petur hugs the h3x0 bootloader usb mode
01:28:06linuxstb_toffe82: It's about time you flashed Rockbox on the gigabeats then...
01:28:11toffe82I experience it yesterday, I fix it using an image of one working
01:28:46linuxstb_How did you write the image? Do you have to attach the HD to a computer in another way?
01:28:57toffe82linuxstb_: just waiting the jtag interface, will have it next week, was to lazy to make one ;)
01:29:47toffe82when you put a new hd, the system says "no system found", if you connect the usb, you have can see it from your computer
01:29:57 Part straylight42
01:30:00jhMikeSwell, how do we make a special program of our own then?
01:30:19toffe82what do you mean ?
01:30:38jhMikeSyou said if you put a new drive in the gigabeat it needs a special program?
01:30:41linuxstb_Can't we just work out what needs to be written to the disk?
01:30:52JdGordonanyone with an ipod got a min to test this patch? Im not 100% sure i got the logic right :p
01:31:17toffe82we could find what need to be written
01:31:49toffe82jhMikeS: I used an image to reload it
01:32:16jhMikeSso you have a dump of a working drive?
01:32:23linuxstb_But it sounds like Toshiba's firmware needs to be banished to hell anyway...
01:32:32preglowjhMikeS: want me to tell you what think of having to waste two regs on an accumulator?
01:32:35toffe82no just an image made with acronis drive image
01:33:06*preglow kicks arm in the teeth
01:33:16preglowand i know that sounds anatomically incorrect
01:33:18jhMikeSpreglow: I'm a ways from that experience
01:33:18JdGordonarms have teeth?
01:33:28toffe82going home, talk to you later
01:33:34preglowjhMikeS: doing downsampling now
01:33:38 Part toffe82
01:33:53 Join Thundercloud_ [0] (n=thunderc@
01:34:03jhMikeSI'm sure I'll not be happy about it
01:34:56JdGordonsurely someone wants to test a quick path before its commiteD?
01:35:02jhMikeSOn cf you can use the accumulators to just store data for later instead of pushing to the stack if you like :)
01:35:13preglowjhMikeS: and it's wonderful!
01:35:40JdGordon <- set start screen to wps and the setting so it doesnt resume if headphones arnt connected to on and reboot
01:36:40 Quit JdGordon ("CGI:IRC")
01:36:49 Join JdGordon [0] (i=82c20d65@gateway/web/cgi-irc/
01:38:32DataGhostshame on me, I lied... I wasn't going to bed
01:38:36jhMikeSThe screen rotation thing has to go in for sure. The gb LCD is sharp and clear from any angle.
01:38:39DataGhostat least it reads the thing semi-correctly
01:39:13 Join midkay_ [0] (
01:39:26JdGordonDataGhost: whats that? the ata stack?
01:39:28jhMikeSlogf doesn't need newlines :)
01:39:43*JdGordon sees lots of DEADBEEF and gets hungry :p
01:39:46linuxstb_jhMikeS: How could you deal with LCD_WIDTH and LCD_HEIGHT no longer being constant?
01:39:54DataGhostthat's supposedly my FAT32 superblock
01:39:58DataGhostbut it looks awfully short
01:40:05DataGhostalmost as if it's an weirdly-aligned read
01:40:16petursuperblock? you mean MBR?
01:40:18DataGhostnot 100% sure though, it could also be my function not working properly
01:40:22DataGhostfat header
01:40:28jhMikeSlinuxstb_: I'm not sure they'll need to change or just change things so it doesn't have to be.
01:41:24linuxstb_The choice of which .bmp files to compile into Rockbox is based on LCD width and/or height for example.
01:42:14DataGhostit's reading it reversed
01:42:16Shaidahh, good ol' dead cow.
01:42:33DataGhosteb58906d is on my disk
01:42:43DataGhost6d9058eb is what it reads
01:42:44jhMikeSWhy change the the bmp defs?
01:42:49peturDataGhost: AA55 is the end of the first sector (bootsector)
01:43:00DataGhostit's 55AA
01:43:06jhMikeSIs the rotation supposed to be general for all UI in current form?
01:43:09DataGhostas I just said, it's reading everything reversed :)
01:43:25DataGhostbut for fat32 that ending is the same :)
01:43:40JdGordonDataGhost: endianess problem?
01:43:46DataGhostlooks like it
01:43:50DataGhoststrange though
01:44:10linuxstb_jhMikeS: The only patch I've seen permanently rotates the LCD, but everyone seems to be requesting it as a setting.
01:44:30*JdGordon wants it as a setting on the h300 also!
01:44:56linuxstb_It would work well on the H10 20GB - a nice 128x128 LCD...
01:45:09jhMikeSIt shouldn't be hardcodec nor should say, the USB logo be doubled up
01:46:10jhMikeSYou put in square bitmaps that get clipped on the shorter axis
01:46:46JdGordondoes anyone know what cast is needed for int -> void* for the 64bit comps?
01:47:21JdGordonisnt that for pointer > int?
01:47:24jhMikeSshould really use that instead of void parameters
01:47:29 Quit entheh ("^~")
01:47:29jhMikeSeither way
01:47:34JdGordonoh ok
01:48:55preglowJdGordon: you really should use intptr_t instead of void *
01:49:18 Quit Thundercloud (Read error: 113 (No route to host))
01:49:24preglowthat way you can forget about the problem altogether
01:49:26*jhMikeS changes all the queues and codecs already :)
01:50:19jhMikeSThe void * thing makes code very messy
01:50:28 Quit rotator ()
01:50:38DataGhostthis is going to be the hell to decode
01:50:49DataGhostit's almost as if it's reading byte-random from my superblock
01:51:16DataGhost66730000... you'd expect to see 00007366
01:51:18DataGhostbut nooooo
01:51:30Shaiddoes the 5g ipod support recording via the headphone port?
01:52:37preglowjhMikeS: sometimes i also miss the ability to do constant offsets in addition to indexed and scaled offsets
01:52:47preglowthat is, at the same time
01:54:18 Quit JdGordon ("CGI:IRC")
01:54:40preglowthis'll still throw warning...
01:54:51 Quit midkay (Read error: 110 (Connection timed out))
01:55:22 Join Thundercloud [0] (n=thunderc@
01:56:16 Quit midkay_ (Read error: 104 (Connection reset by peer))
01:56:41 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
01:57:57preglowjhMikeS: really, should the downsample loop ever be done before it has even done one iteration?
01:58:28 Join midkay [0] (n=midkay@rockbox/developer/midkay)
01:59:26preglowand the "nice fix" award goes to jdgordon
02:00:54LloreanThe 'fix yellow' fix that greatly increases the amount of yellow?
02:01:28 Quit Thundercloud_ (Read error: 145 (Connection timed out))
02:01:31 Quit p0ser (Remote closed the connection)
02:01:39preglowwell, there's more yellow than green now
02:01:43preglowmight be interpreted as fixing
02:01:54LloreanIt fixes the lack of yellow, I suppose.
02:02:31jhMikeSpreglow: yes
02:02:46jhMikeSit should adjust the phase and not output anything
02:03:42linuxstb_Llorean: Did you notice that I've now wiped all ipod install instructions from the wiki, replacing with references to the manual?
02:04:16Lloreanlinuxstb_: Good to hear.
02:04:19linuxstb_Linus also removed all the old files from the download server, so external tutorials will now be broken.
02:04:26preglowjhMikeS: you do things in a different order in the asm versus the c version, btw. in asm, while loop condition is checked first, and then last sample is updated if pas > 0
02:04:48Lloreanlinuxstb_: That's very good to hear.
02:05:22preglowin c, pos > 0 and last sample update is first checked, then while loop condition asm, the bne goes to load both the previous and last if pos != 0
02:09:07jhMikeSif pos == 0, the current sample is loaded into %d1 and the last sample saved is already in the struct
02:09:13jhMikeSin %d0 rather
02:10:29jhMikeSI just have finer branching control in asm afterall
02:11:47preglowthe if (pos < count) check will branch to .dsloop_skip before the if (pos > 0) check can be taken in the asm
02:12:00preglowresulting in last_sample not being updated for that frame if count is bigger than pos from the very start
02:12:38jhMikeSlast_sample is always updated into resample_data
02:14:48preglowtell ya what
02:14:50preglowif you just ignore me
02:14:51jhMikeS move.l -4(%a3, %d2.l*4), (%a5) writes it before any branching is done, %d0 always gets the previously saved value
02:14:53preglowi'll promise to go to bed
02:15:15preglowyeah, nothing to see here
02:15:16preglowmove along
02:15:22preglowturns out i'm too tired for this
02:15:29preglowso think i'll toddle off for today
02:15:40jhMikeSgoodnight then :)
02:15:47preglownightie :/
02:16:29DataGhostI think I know what's wrong and I'm beginning to feel quite stupid now
02:17:00DataGhostI know
02:17:04Shaidalways a good feeling
02:17:11 Join Hoffmann [0] (
02:17:18DataGhoststupid assumption :)
02:17:36Shaidall assumptions are stupid.
02:17:40Shaidor so I assume. ;)
02:19:03DataGhostthere's actually a huge difference in addressing, usage and interpretation of an int(32) instead of a char
02:19:08 Quit actown (Read error: 104 (Connection reset by peer))
02:19:11*DataGhost whistles innocently
02:19:58DataGhostgood god.
02:25:07DataGhostno more endianness problem \o/
02:25:25DataGhostI'll get better pics tomorrow, this was the last picture my cam wanted to make
02:26:52 Join Brunellus [0] (n=luigi@unaffiliated/brunellus)
02:27:02 Part Brunellus ("Ex-Chat")
02:27:24 Join Noah0504 [0] (n=noah@
02:28:51Noah0504If I format an iPod will it remove the default Apple OS?
02:29:22LloreanNoah0504: It depends, to an extent, on how you format it.
02:29:52Noah0504Well, let's say I want to remove the Apple OS, haha. I just want a nice clean iPod running Rockbox.
02:29:57LloreanNoah0504: If you format the entire physical drive, it'll destroy the partitions, and you'll lose the firmware partition. If you just format the FAT32 partition, it'll be fine.
02:30:54LloreanAt the moment the best way to remove the apple OS is to -d the current bootloader, -wf a new bootloader, and then resize and shrink the firmware partition.
02:31:10[Tesser]Noah0504- You barely notice the presence of the original firmware. Why go to extra trouble?
02:31:31Llorean[Tesser]: The original firmware partition is 40-120mb, plus it takes longer to boot up while it's present.
02:31:59Noah0504[Tesser]: I was thinking about formatting anyway, and I was wondering if that would work.
02:32:14LloreanYou can cut 1-2 seconds off the boot time, and recover a decent amount of disk space if you remove it.
02:32:22 Join toffe82 [0] (
02:32:33[Tesser]Hmmm. Interesting.
02:33:04[Tesser]Those sound like good reasons.
02:33:25LloreanIt's most noticeable (for the boot time thing) on the Nano.
02:33:44LloreanOn the disk based iPods, there's a spinup time on the disk, that makes up a decent enough chunk of the boot time.
02:34:11[Tesser]I hardly notice on my 5.5
02:34:15Noah0504Well, I think I'm going to be formatting under Linux using mkdosfs. If I just specify the device, will it format it entirely, or do you not know?
02:34:35Llorean[Tesser]: You can miss it if you blink on my Nano, set up properly.
02:35:01LloreanNoah0504: I don't think it's a good idea to format the whole device. You _need_ the firmware partition there, no matter what.
02:35:03[Tesser]Noah0504 : What type of iPod?
02:35:10LloreanNoah0504: Unless you plan to recreate the firmware partition after.
02:35:51Noah0504[Tesser]: 1st Gen Video iPod.
02:36:03Noah0504Llorean: Ahh, okay.
02:36:16[Tesser]5g then?
02:36:17 Quit petur ("Zzzz")
02:36:26Noah0504[Tesser]: Yeah.
02:36:41[Tesser]Not the fuckety 5.5g with the 'larger' sector size.
02:36:46 Quit Hoffmann (Read error: 110 (Connection timed out))
02:36:47LloreanNoah0504: If you really want to reformat, reformat the Fat32 partition. Then use the ipodpatcher tool with "ipodpatcher -d" to remove the old bootloader. Then "ipodpatcher -wf bootloader-video.ipod" to write the bootloader as firmware.
02:37:11LloreanNoah0504: Then shrink the firmware partition, it shouldn't need to be very large at all.
02:37:21LloreanActually, you may want to save reformatting the Fat32 partition until this point.
02:38:57 Join Hoffmann [0] (
02:39:05Noah0504Well, if I do all of that, I might do it on my Windows machine. Oh, and I did try formatting it once under Windows be just right clicking on the device, but I noticed that it just removed my Rockbox files and not the Apple files. I'm guessing those were recreated by the firmware?
02:39:33 Quit inversions (Read error: 110 (Connection timed out))
02:39:33LloreanThe Apple files get recreated each time you boot the Apple firmware, so long as it's still there on the firmware partition.
02:40:27Noah0504Okay, well, I think I'll do what you told me. Does a smaller firmware partition create a faster boot also? I might not mess with resizing it.
02:40:39LloreanIt's not necessary to resize it.
02:40:46LloreanThat's only really useful for freeing up some additional disk space.
02:40:57LloreanOn a disk based ipod, it's such a small amount relative to the total it doesn't matter much.
02:41:13Noah0504Okay, well, I think I'm going to go give everything a try. Thanks a million for the help.
02:41:21LloreanIf you just do the ipodpatcher -d, and ipodpatcher -wf bootloader-video.ipod (you'll of course need the bootloader-video.ipod file), you'll get the increased boot time.
02:41:52LloreanNoah0504: All the files you need are here:
02:42:10Noah0504Good stuff. I'll be back in a bit!
02:42:18 Part Noah0504
02:48:00 Join linuxipod [0] (
02:48:17linuxipodcan somebody give me a link on how i can put rockbox on 5g ipod?
02:50:13linuxipodor an easy installer
02:50:43safetydanlinuxipod, have you even looked at the website and the link to the manual?
02:51:29linuxipodis there an easy installer
02:51:35safetydanSo... the big section on installing wasn't helpful?
02:52:56linuxipodyes but i was wondering if there was an eaiser installer
02:54:09safetydanIt's pretty easy as is. Download two files, unzip one to your iPod, run the other.
02:56:49 Join barrywardell [0] (
02:59:29 Join lucas42 [0] (i=proxyuse@
03:00:08Lloreanlinuxipod: You double click on a single.exe, and then extract a .zip to your iPod. How much easier do you want?
03:00:53linuxipodohh looking throught the guide now
03:02:15linuxipodOpen up a terminal window and type the following commands:
03:02:18linuxipod cd $HOME
03:02:18linuxipod chmod +x ipodpatcher
03:02:21linuxipod ./ipodpatcher
03:02:21linuxipodwhats that?
03:03:33LloreanAre you on a linux computer?
03:03:51linuxipodno pc
03:04:02Llorean"PC" means "Personal Computer"
03:04:09LloreanDo you mean Windows?
03:04:09linuxipodlol windows
03:04:16LloreanFollow the Windows instructions then.
03:04:24LloreanThose aren't them.
03:04:25linuxipod(those mac, pc commercials)
03:04:33 Join p0ser [0] (
03:04:45 Join kaek_ [0] (
03:04:49linuxipodwhere are the windows instructions
03:05:01safetydanlinuxipod, about two paragraphs up from where you're looking
03:05:09safetydanSee "Bootloader installation from Windows"
03:05:35safetydanwell, more two sections rather than paragraphs, but anyway
03:06:02 Quit p0ser (Client Quit)
03:06:05 Join Noah0504 [0] (n=noah@
03:06:32Noah0504Llorean: Are you there?
03:06:36linuxipodahh thx (Y) i think ive installed it
03:06:39LloreanNoah0504: Yes.
03:06:53 Quit Thundercloud (Remote closed the connection)
03:07:14linuxipodMenu+Select (where is the select button?)
03:07:38LloreanThe center button.
03:08:00linuxipodk im pressing it but nothing happens
03:08:11 Join MrMuffin [0] (
03:08:30linuxipodthe little apple lit up but thats it
03:08:30Lloreanlinuxipod: Then you're doing it wrong. You need to press and hold both buttons for several seconds.
03:08:36LloreanLittle Apple?
03:08:47linuxipodyes in the middle of the screen
03:08:55LloreanThen you rebooted your iPod.
03:09:00***Saving seen data "./dancer.seen"
03:09:09LloreanBut if it just goes back to the Apple software, you didn't follow the installation instructions properly.
03:09:34LloreanPlease, just read the manual carefully. From the fact that you got confused and read the Linux install, I think you may be rushing things. Take your time, and read carefully.
03:10:45Noah0504I need to write one of the .bin found under as the firmware, right?
03:11:18linuxipodok that was wierd i hit the hold switch and rockbox bootlaoder 1.0 came up
03:11:58linuxipodthen it says loading linux... error cant load linux.bin
03:12:02 Join p0ser [0] (
03:12:53linuxipodwhat am i doing wrong
03:13:04LloreanNoah0504: As I said before, yeah, ipodpatcher -wf bootloader-ipodvideo.ipod
03:13:08LloreanNoah0504: Not a .bin file
03:13:22Lloreanlinuxipod: Were you holding Play down?
03:13:49LloreanAlso, there's no reason the hold switch should be on
03:13:52LloreanThe directions don't say to do that.
03:14:00LloreanWhy aren't you just following them?
03:14:39linuxipodall the directions say is "Remove your player from the computer’s USB port. Rebooting the Ipod by holding Menu+Select for a couple of seconds until the player reboots. Now Rockbox should load. When you see the Rockbox splash screen, Rockbox is loaded and ready for use."
03:14:53linuxipodnothing loads
03:14:55LloreanSo why did you turn hold on?
03:15:17linuxipodi was just seeing what would happen
03:15:18LloreanIf you aren't seen the Rockbox splash screen when you do that (if it goes straight into Apple) you did something wrong.
03:15:33LloreanDid you "just see what would happen" at any other step along the way?
03:15:41Noah0504Llorean: Sorry, that's what I meant. I removed the Apple firmware and wrote the Rockbox bootloader, however I forgot to extact the Rockbox firmware before I unplugged the iPod. Now I just get the glowing Apple when I turn on my iPod, and I don't know if it's going to do anything if I plug it in.
03:15:52Lloreanlinuxipod: And, did you "Eject" the iPod, or did you just pull the plug?
03:16:29LloreanNoah0504: If you just get the glowing apple logo and nothing else, something went wrong. You should be able to reboot into disk mode with Menu+Select then Select+Play though.
03:16:39 Quit barrywardell ()
03:17:33Noah0504Llorean: Hmm. Any ideas of what to do?
03:18:28LloreanNoah0504: Once you connect it in disk mode, use ipodpatcher to get the status of your iPod. I can't remember the exact command. How big does it say your firmware image is?
03:18:29Noah0504Llorean: Wait, nevermind. Let me see something...
03:19:22linuxipodyes i ejected and yes i waited
03:19:29 Quit pixelma (Read error: 110 (Connection timed out))
03:20:39Lloreanlinuxipod: I didn't ask anything about waiting. I asked if you did anything else to 'just see what would happen'
03:21:03linuxipodi dont understand what you mean by that
03:21:39Lloreanlinuxipod: They're your own words. You tried something random without asking earlier, something not in the instructions. Is it possible you did anything else not in the instructions.
03:22:59Noah0504Llorean: Know where I can find the ipodpatcher commands?
03:23:02linuxipodohhh i see no i do not think soo...( i pluged in the ipod, ran the ipodpatcher, ejected the ipod, and then it went to the apple, and i waited there, then i clicked the hold button and it told me that there was an error loading rockbox
03:23:10kaek_Hi, been using a jukebox studio 20 for _years_ with Rockbox firmware, but under a Windows environment. HDD died. Installed Ubuntu, not wanting to pay the MS tax. Now I can't seem to get the thing to connect. Various tests show that the OS is finding and loading the driver for the ATA Bridge, but not the HDD... any ideas?
03:23:10LloreanNoah0504: ipodpatcher −−help, probably
03:23:38Lloreanlinuxipod: I thought you said there was an error loading Linux.
03:23:46Lloreanlinuxipod: You need to tell me exactly, accurately, what it said.
03:24:05Lloreanlinuxipod: I cannot read your mind, and if you tell me the wrong things, I'm not going to be able to help you at all.
03:24:06linuxipodkk hold
03:24:10linuxipodim sry
03:24:22linuxipodill try to be more spacific
03:24:33LloreanError messages have specific meanings.
03:24:59LloreanRockbox's loader CAN load iPodLinux, and if it's trying to, there's something strange going on. If it said it couldn't find Rockbox.ipod, something entirely different is going on.
03:25:10Noah0504Llorean: Okay. Main firmware: 47112 bytes and RSRC: 5242880 bytes
03:25:27Lloreanlinuxipod: If it can't find rockbox.ipod, it sounds like you never extracted on to your device. Which means you didn't read all the instructions. Something I told you to do more than once.
03:25:37LloreanNoah0504: How big is your bootloader-ipodvideo.ipod?
03:26:01 Part kaek_ ("Leaving")
03:26:22Noah0504Llorean: 46KB
03:26:40LloreanNoah0504: I was hoping for the actual number of bytes. ;)
03:26:49Noah0504Give me one sex.
03:27:08linuxipodyour right i did not extrac anything to my ipod
03:27:11Noah050447120 bytes
03:28:05Lloreanlinuxipod: It is a good plan to read ALL the directions, as I told you. You skipped entirely section 2.2.2, which is actually entitled, I believe, "Installing the Firmware", something that would usually be of interest to one who wants to install Rockbox
03:28:07linuxipodwhere do i get the firware from
03:28:19LloreanJust read the manual, PLEASE
03:28:21LloreanIt tells you these things
03:29:02LloreanNoah0504: Hm.
03:29:20Noah0504Should I just try removing it and giving it another shot?
03:29:31linuxipodwhen i got to i dont know what to do (there are not release builds for the 5g ipod)
03:29:35LloreanNoah0504: Actually, when the Apple logo is on the screen, try tapping Right a few times.
03:29:51Lloreanlinuxipod: Read the page.
03:29:57Lloreanlinuxipod: There's text at the very top of it.
03:31:28 Quit MrMuffin ()
03:33:15Noah0504Llorean: I installed the firmware and it seemed to take care of everything. It boots into Rockbox that is...
03:34:06LloreanThat's actually what I was thinking it might be, when I asked for you to tap right (would've given some output telling what was up)
03:34:47Noah0504Llorean: Well, it boots faster, and I removed the Apple firmware, so I guess goal accomplished. Thanks a million.
03:35:12LloreanGood, good.
03:36:41 Join lukaswayne9 [0] (
03:38:53linuxipodLlorean: I think i got it to boot. Now its on this screen that has like file, database, resume playback, and others
03:39:07Lloreanlinuxipod: Yes, that's the main menu.
03:39:30linuxipodthats cool thx
03:42:07 Quit lukaswayne9 ("Ex-Chat")
03:42:20linuxipodhow do i exit a game?
03:42:28linuxipodor (plugins)
03:42:57LloreanI'm not really sure. For a lot of them it used to be Menu+Select, but I use my iPod for music so I haven't been in one in a while
03:43:02LloreanA _lot_ of answers are in the manual
03:43:05LloreanDid you look there?
03:48:40linuxipodthere is no manual for that
03:52:17 Join winchester [0] (n=winchest@
03:52:48SirFunkhey, what was the flac encoder that was fast?
03:54:09linuxipodwhat do i do if it freezes
03:55:17Lloreanlinuxipod: What froze?
03:55:58LloreanI meant, what were you doing?
03:56:01linuxipodohh i just rebooted it ( what does it mean database not ready
03:56:16Llorean"database not ready" means the database isn't ready. You haven't initialized it.
03:56:22LloreanSeriously, try looking things up in the manual
03:56:24LloreanThey're there.
03:56:33linuxipodim looking in the manual
03:56:51LloreanIf you just type "database" into a search of the PDF manual, you'll get plenty of information on it.
03:58:31 Join bombi_ [0] (
03:58:35winchesterHeya.... what's with rockbox freezing with a "Loading..." splash on start-up sometimes? (or is that just me?)
03:58:39winchester(on an ipod)
03:59:27Lloreanwinchester: What type of iPod?
03:59:47winchester2g mini and... tha happens when starting while the USB cord is connected
03:59:57winchesterin about 1/5 times
04:00:01LloreanYes, there are problems starting with a USB cord attached right now.
04:00:17winchesterhmm ok
04:01:53winchesterand... is there some place on the net that has unmodified firmware files?
04:03:03linuxipodthx ill be back tommorow
04:03:06 Quit linuxipod ("Chatzilla 0.9.77 [Firefox]")
04:03:45Lloreanwinchester: Unmodified firmware files?
04:04:00LloreanYou can uninstall without needing one.
04:04:57winchesterbut... if I have Loader 2, that wouldn't exactly work would it?
04:05:45Lloreanipodpatcher -d should remove loader 2 as long as it's properly installed.
04:06:11LloreanThen again, you shouldn't be asking for any help with Loader2 here.
04:06:40LloreanAs well, if it's freezing while booting with loader2 while showing loader2's "loading" message, your problem probably isn't even caused by Rockbox.
04:07:06winchesternuh.... definatly Rockbox.... the logo int he background with the rockbox API splash
04:08:09LloreanRockbox does not have a screen that says "Loading"
04:08:30LloreanNot normally, at least.
04:08:37winchesterexactly.... only comes up when that happens
04:10:08 Quit bombi_ ("Verlassend")
04:10:13LloreanWell, if it has the Rockbox logo in the background, there's a decent chance it's a Rockbox screen then
04:10:18LloreanHave you tried it with the official bootloader?
04:11:27winchesterNo, not that concerned about it.
04:13:29 Join dsh-1 [0] (
04:14:22LloreanIt'd be kinda nice to know more about the problem, but most people who've mentioned it haven't been using the up to date Rockbox software, and are often using strange combinations like you that may affect things.
04:17:35winchesterI'll see if it still happens without loader 2 then
04:18:14 Part Llorean
04:21:30winchesternope.... still there.
04:31:45winchesterhey, does ipodpatcher -wfb copy just the apple firmware or the whole partition?
04:33:26 Part safetydan
04:34:05 Join BHSPitLappy [0] (
04:39:53 Join rotator [0] (n=e@rockbox/developer/rotator)
04:40:10 Join Shaid` [0] (
04:41:07 Quit Mouser_X (Nick collision from services.)
04:41:16 Join Mouser_X [0] (
04:42:51 Quit winchester ("Leaving")
04:49:27 Join Shaidd [0] (
04:49:35 Quit Shaid (Nick collision from services.)
04:49:39 Nick Shaidd is now known as Shaid (
04:52:33 Join JETC- [0] (
04:54:49Noah0504Ooo, the latest builds of Rockbox are weird. Took me a good second to figure it out.
05:03:09 Quit BHSPitLappy (Remote closed the connection)
05:08:00 Quit Shaid` (Read error: 110 (Connection timed out))
05:09:01***Saving seen data "./dancer.seen"
05:11:34perldivermenu.c: In function 'menu_show':
05:11:34perldivermenu.c:702: warning: initialization makes pointer from integer without a cast
05:14:27 Quit lucas42 (Read error: 110 (Connection timed out))
05:22:22 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
05:23:52 Part jhulst ("Kopete 0.12.4 :")
05:24:04 Join linuxipod [0] (
05:24:19linuxipodis there a way to get back to normal ipod firmware (without removing the rockbox)
05:26:37Noah0504linuxipod: Reboot the iPod by holding down MENU and SELECT at the same time.
05:26:58linuxipodthen it will boot normally
05:26:59Noah0504Then hold down MENU until you see "Loading original firmware..."
05:27:36linuxipodahh thx
05:27:42Noah0504No problem.
05:27:43 Join Hoffmann_ [0] (
05:31:27 Join BiptoN [0] (i=4ca6c5a0@gateway/web/cgi-irc/
05:32:35BiptoNdoes the broadcom chip in the 5g and 5.5g ipods drive the lcd or only video?
05:34:00 Quit Hoffmann (Read error: 110 (Connection timed out))
05:36:16 Join aliask [0] (
05:36:43Noah0504BiptoN: I want to say the video, but I could be wrong.
05:42:51SirFunkwhat is the flac encoder that makes bit-compatible flac files, but is faster than normal flac?
05:44:21 Quit Hoffmann_ (Read error: 110 (Connection timed out))
05:45:43 Join Hoffmann [0] (
05:47:37 Quit p0ser (Client Quit)
05:48:52FebsSirFunk: FLAC 1.1.4, which was released a couple of weeks ago, is significantly faster than prior versions.
05:49:43BiptoNi wonder if putting the broadcom to sleep would save power on the ipods, if it does only handle video
05:53:34 Quit perplexity_ (Remote closed the connection)
05:55:00aliaskBiptoN: Why, do you know how to? :)
05:55:42SirFunkFebs: i thought there was another encoder
05:55:44SirFunkflak or something?
05:56:51BiptoNwell i am not real bright
05:57:06BiptoNand was wanting to know if it drove the lcd first
05:57:18BiptoNi'll be lookin for the info on it now
05:58:08BiptoNi get 15 hours or more constant with the bummer apple software and rockbox only hits 8 to 9
05:58:42BiptoNthere has to be a chip or two consuming power when it needs not to
05:58:53BiptoNlike the x5 usbtg chip
06:00:35aliaskI think you could be right in guessing that the broadcomm chip is sucking a bit of power, but the problem is we don't know how to do ANYTHING with that chip, even turn it off.
06:00:48aliaskOr at least, that's my understanding of the situation.
06:01:19BiptoNyeah that's what i thought, i'm gonna try scourin for docs on it, and ask a few people if they have any access to em
06:01:38aliaskI wish you the best of luck.
06:04:22 Join Hoffmann_ [0] (
06:22:22 Quit Hoffmann_ (Read error: 110 (Connection timed out))
06:22:24 Join fejfighter [0] (
06:24:37 Quit Hoffmann (Read error: 110 (Connection timed out))
06:25:56 Join Hoffmann [0] (
06:26:21 Join Aaron4 [0] (
06:28:15 Part Noah0504
06:29:32 Quit dsh-1 (Remote closed the connection)
06:30:49 Join Llorean [0] (
06:31:25LloreanBiptoN: The Broadcom chip drives the LCD, we have to use it for LCD updates. Plus, all other iPods suffer battery life reductions of similar ratios to the 5G suggesting it's hardware in all of them (something in the portalplayer chips, a good guess would be USB hardware again)
06:34:15BiptoNwhat about firewire?
06:34:27BiptoNthere has to be something
06:34:42LloreanThe 5G doesn't have firewire.
06:34:53BiptoNafter the KoC patch my 4g gained 2 hours play time
06:35:03LloreanAs I said, a good guess would be the USB hardware again.
06:35:16BiptoNis it the lack of info about the phillips power management chip?
06:35:32LloreanStop for a second.
06:35:37LloreanIf we knew what it was, we'd know what to research.
06:35:58LloreanSo it's kinda silly asking me "Is it X" because the answer is going to always be "We don't know what it is yet."
06:36:12BiptoNgood call
06:36:17LloreanBut common sense suggests very strongly that it's not the Broadcom chip.
06:36:44LloreanIt also suggests less strongly, but still decently, that since the USB On the Go hardware caused problems in the H300 and X5, there's a decent chance it could be causing problems in all the iPods.
06:36:47 Join _fejfighter_ [0] (
06:36:57BiptoNi'm gonna do what i can to find out if there are any datasheets in other folks hands
06:37:14LloreanThere is no guarantee that's the problem, but since it is undocumented hardware, and in a category of hardware that has been a past problem, it'd be a good start.
06:37:26BiptoNalright Llorean
06:37:40LloreanThen again, we have information because it's related to the... was it i.MX3?
06:37:52LloreanSome USB controller that does have a datasheet, or at least more information.
06:38:03LloreanSo there's at least a little more confidence that we're handling that right.
06:38:47BiptoNso datasheets on the usb hardware would help out than
06:39:11BiptoNhas anyone emailed nvidia to request portalplayer docs?
06:39:30 Quit desowin ("use linux")
06:39:42LloreanNVidia doesn't have a tradition of helping open source, and PortalPlayer itself did not even bother to respond to emails with a rejection.
06:42:49 Quit Hoffmann (Read error: 110 (Connection timed out))
06:42:56 Join Hoffmann [0] (
06:44:56 Quit Febs (Read error: 110 (Connection timed out))
06:48:49 Quit fejfighter (Read error: 110 (Connection timed out))
06:51:37 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
06:51:41 Part toffe82
06:56:14 Part Llorean
06:56:38SirFunkthat flac encoder i was looking for is called
06:56:42SirFunkthat took me forever to find
06:59:16crwli quickly tried it once, but it couldn't write tags to the encoded file then, so i couldn't use as a drop-in replacement to flac and gave up
06:59:42 Quit Hoffmann (Read error: 110 (Connection timed out))
06:59:47SirFunkwell i use wavpack now
06:59:54SirFunkbut i'm going to try some benchmarks
07:05:00crwlif your ripper can do the tagging, that won't matter... kaudiocreator can't, it can just pass some arguments to the encoder
07:09:05***Saving seen data "./dancer.seen"
07:09:08 Quit JB_Away (Read error: 131 (Connection reset by peer))
07:09:30 Join JB_Away [0] (
07:14:21 Join Larsie [0] (
07:14:33 Quit Larsie (Client Quit)
07:21:45 Quit rotator ()
07:22:29 Join decayedcell [0] (i=3ba79d47@gateway/web/cgi-irc/
07:22:53 Join decayedcell_ [0] (i=3ba79d47@gateway/web/cgi-irc/
07:26:26*amiconn wonders what JdGordon is doing
07:27:29 Quit decayedcell ("CGI:IRC (Ping timeout)")
07:43:21 Quit decayedcell_ ("CGI:IRC (EOF)")
07:46:05 Join decayedcell__ [0] (n=decayed_@
07:46:20 Quit Aaron4 ()
07:57:02 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
07:57:15 Join miepchen^schlaf [0] (
07:57:47 Join pondlife [0] (
08:00:16 Quit midkay ("Leaving")
08:18:50 Join eggy [0] (i=eggy@pdpc/supporter/sustaining/yorkcc.eggy)
08:19:48 Join blueworm [0] (
08:21:05 Quit decayedcell__ (Remote closed the connection)
08:26:01 Join Rob2222 [0] (
08:27:08 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
08:33:05DataGhosthow's that eh? :P
08:33:24DataGhostnow I just need to sort the fsinfo and see if it'll actually mount it
08:35:19*eggy wonders
08:37:22jhMikeSat least you're getting somewhere with what you're doing :)...can't say the same for myself regarding database and recording both writing to disk
08:37:31eggythe coLinux maintainer dude around?
08:38:38 Join bluebrother [0] (i=QCvz5KRj@rockbox/staff/bluebrother)
08:41:30DataGhost yaay :)
08:41:35bluebrotherB4gd3r: the m5 manual builds and can get added to the daily manual page
08:43:13 Quit Rob222241 (Read error: 110 (Connection timed out))
08:43:41 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
08:44:01 Join decayedcell__ [0] (n=decayed_@
08:48:55 Quit GodEater_ ("CGI:IRC")
08:49:15pondlifejhMikeS: No luck yet?
08:49:50 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
08:50:24jhMikeSpondlife: there's no lock. if you wait, the flush will finish. so that's good at least
08:51:05pondlifeHow long does it take?
08:51:18GodEater_DataGhost: what's that a screenshot of ?
08:51:23pondlifeA normal database init takes about 8 mins here
08:51:24jhMikeStook about 20min...but I just had a breakthrough now
08:51:35pondlifeI left it for 15 mins with no movement
08:51:56GodEater_DG: on which model ? :)
08:52:02DataGhost80GB :)
08:52:18GodEater_so you've given up waiting for LinusN to invent timetravel too huh ? :)
08:52:32 Join kubiix [0] (
08:52:33DataGhostjust thought I could maybe apply some of my new knowledge
08:52:43GodEater_I'm sure he'll be very grateful
08:53:07GodEater_as will we all be that own the 80G
08:53:19pondlifeI'd like to own an 80G one day
08:53:20 Join KWha1 [0] (
08:53:22jhMikeSThe low priority thread calls yield with the ata mutex locked...following? yields inside the mutex, but doesn't get a chance to run and discover the drive is ready for read/write and blocks the higher priority thread from completing its write
08:53:38GodEater_I thought I didn't recognise the "This isn't FAT32 go away!" message as being part of the linux kernel
08:54:02jhMikeSI denied the yield to background threads and the flush completed in a timely manner
08:54:22KWha1question about themes, just installed rockbox and going through the manual. I figured out how to change themes but they are all black and white like they are missing the images or color info... ideas?
08:54:44jhMikeSI will verify this again though and see if I have to eat all those words :)
08:55:31amiconnjhMikeS: Which thread is low priority during disk write??
08:55:33jhMikeSwhat are the files created by database so I can delete them? just one
08:55:44jhMikeSamiconn: the tagcache thread
08:55:48GodEater_Llorean: I discovered this morning that grip (linux front end to cd ripping / encoding) has a nice "Wav filter" feature, letting me running sox automatically on each track after it's ripped, but before it gets passed to the speex encoder
08:57:24amiconnGrip is very nice and I use it when ripping on linux, but some of its defaults are insane imho
08:57:36GodEater_Llorean: oh, and I logged the fact that speex stereo doesn't play back in rockbox yet as a bug on flyspray.
08:58:07GodEater_amiconn: do you mean it's ripping defaults? or encoding ?
08:59:09jhMikeSwe might need a special yield that guarantees thread execution of the thread on every round for these shared resources...
08:59:24jhMikeSshure as heck that stops it :)
08:59:38amiconnGodEater: File name generation.
09:00:38amiconnWith default settings, it makes all files lowercase, replaces spaces with underscores, and strips all sorts of characters which are valid in filenames
09:01:07jhMikeSanything holding that mutex basically has to temporarily be realtime
09:01:31KWha1anyone have any idea why theme graphics would not be showing up after a fresh install
09:01:37amiconnjhMikeS: That might also cause problems
09:01:59DataGhostI just crashed rb
09:02:06DataGhostafter getting a succesful mount and disk reads :P
09:02:14jhMikeSnot in the sense of denying other thread, but just be guaranteed to not be denied rounds
09:02:40amiconnYes, but that can also cause ugly problems
09:02:43jhMikeSdifferent than a thread priority of realtime
09:02:45jhMikeShow so?
09:03:09amiconnOn ipod, playback of some formats is really tight, and its better to let the rebuffering wait a bit longer than let playback skip because the codec doesn't get enough cycles
09:03:23amiconnAnd rebuffering does access the disk
09:03:46jhMikeSwe wont deny many cycles cause the thread will check and quickly yield again
09:03:54amiconnThat was one reason why this priority thing was introduced
09:04:08amiconnHmm, this probably needs testing
09:04:44amiconnThis scheduler is too complex for being able to correctly predict what will happen :/
09:05:05DataGhostbut it crashes after about 30 seconds
09:05:18decayedcell__hmm has the Database navigation mode been changed?
09:05:40decayedcell__When I'm playing a song in Database mode, pressing Select no longer goes back into the artist/album etc tree
09:06:38jhMikeSI think it will fix it and not prevent other threads from timely execution. The disk writes/reads themselves don't yield anyway. We can't have a low priority thread blocking everything out.
09:07:10amiconnThat's true
09:07:13jhMikeSright now it's basically a terrible priority inversion of sorts
09:07:18jhMikeSreally bad
09:07:21amiconnPerhaps mutexes should be part of the scheduler?
09:07:27 Quit GodEater_ ("CGI:IRC")
09:07:33amiconnNot sure whether that would be better...
09:07:47 Join ender` [0] (
09:08:06KWha1can some one help me figure out why theme graphics are not being displayed?
09:08:50jhMikeSI thought about introducing priority of entry in a queued manner, I'm guessing it won't help this anyway.
09:08:56 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
09:09:06***Saving seen data "./dancer.seen"
09:10:12 Part decayedcell__
09:10:28jhMikeSIf the thread can execute in that wait loop so it finds out the disk is ready in a short time, it can complete its access and give the mutex to higher priority threads.
09:10:36DataGhostanyway I still need to do a few other things :(
09:11:04 Join tucoz [0] (n=tucoz@rockbox/staff/tucoz)
09:11:43tucozSlasheri, the audiohw_init() helped
09:12:23tucozSlasheri, now the beep lasts for maybe 1/10th of a second instead of half a second
09:13:53tucozSlasheri, i think the bootloader has to reset the UDA to get rid of the beep completly.
09:14:12 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
09:14:33 Quit zylche ("-")
09:16:20DataGhosthey LinusN
09:16:21jhMikeSSo I suppose the next step is cleanly placing in a special type yield
09:16:39DataGhostLinusN :)
09:17:39jhMikeSAfterall recording on the codec side takes advantage of increasing a thread's priority and yielding more in order to let other things run.
09:17:47LinusNDataGhost: niiiiice - care to share the patch so i can have a look?
09:18:24DataGhostI'm not sure... if I tell you that rockbox crashes after about 30 secs (or when trying to actually open a directory/mp3 or something), do you still want it?
09:18:40DataGhostprobably some cache issue
09:18:44DataGhostbecause I saw it complain about that
09:19:08LinusNthere are plenty of nasty issues with that driver, i can tell
09:19:31DataGhostwait a bit, I'll see if I can reproduce that cache error thingy and take a picture of it
09:19:37DataGhostafter that I'll make a patch
09:19:45DataGhostand you'll get it only if you promise you won't kill me
09:20:13GodEater_LinusN has been known to track people down to the ends of the earth when he doesn't like their code
09:20:18LinusNDataGhost: just curious, have you considered the backward compatibility or just hacked it to work on the 80gb?
09:20:24DataGhostjust hacked
09:20:40DataGhostsame way I did with the linux kernel.. hack first, refine later
09:21:47DataGhosthm and it also gave me ROLO error: something
09:21:56DataGhostwhen I tried to open rockbox.ipod (because that gives me the cache errors)
09:21:58jhMikeSdoes each pp core have it's own irq level? if so, set_irq_level_and_block_thread needs a tweak.
09:23:05 Quit perldiver (Read error: 110 (Connection timed out))
09:23:21DataGhost there we go LinusN :)
09:23:56LinusNodd sector number... figures...
09:24:16tucozDataGhost, you have to insert a bootable floppy in your ipod
09:24:28tucozand press the any key ;)
09:24:30 Join MeKi||a [0] (
09:25:30DataGhosthm I think I only changed fat.c
09:25:37DataGhostnot even ata.c
09:25:42DataGhostthe changes in there are commented out :)
09:33:37pondlifeJdGordon: ping
09:34:29jhMikeShmmm...I think adding the special yield may be straightforward actually
09:36:20 Join roolku [0] (
09:36:50 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:38:05 Join anotherbrick [0] (
09:39:50 Join _Veseliq_ [0] (
09:39:56anotherbrickQuestion, I've just update my build for the first time in about a month an a half on a 30 gb ipod 5g, and recieve a message "data abort at 0003011C" upon playing any mp3s, I haven't tested FLAC or other codecs, nor have I tried an older build since, anyone have suggestions? is this a known issue?
09:40:23 Join mcphail_ [0] (
09:40:41pondlifeanotherbrick: You need to update your bootloader too
09:41:05anotherbrickgotcha, when was the bootloader updated, and what were the changes made to it?
09:41:20pondlifeDon't know, but it's needed for co-processor support
09:41:36anotherbrickalso, since I dont have time to go through that process at the moment, what's the newest build I could roll back to with the old bootloader temporarily?
09:42:28 Quit scorche (Remote closed the connection)
09:42:54pondlifeanotherbrick: 3rd March or older I think
09:42:58 Part KWha1
09:43:37anotherbrickthanks pondlife.
09:43:48safetydananotherbrick, the process of installing the bootloader has also got a lot simpler
09:45:06anotherbrickyeah, but i feel like sleeping now, and I'll be wanting tunes while I work on my senior project all day tomorrow =)
09:45:11pondlifeJdGordon: Not around then? Anyone else know root_menu.c..?
09:45:11anotherbricki'll update that tomorrow night.
09:45:33anotherbrickthx for the feedback safetydan
09:46:10safetydananotherbrick, righto, just letting you know it's not as scary any more :)
09:46:51anotherbrickhaha, for sure, I never found it scary really, just a little time consuming (to back up old bootloader, etc.)
09:47:45peturamiconn: I have to agree that 80GB disk spins up a bit faster, but what I like most is that it also feels much faster loading stuff from disk (like starting rb)
09:48:33jhMikeSpetur: probably is if it spins the same rate
09:49:22jhMikeSnot that I was invited to respond to that :P
09:51:47 Quit _Veseliq_ (Remote closed the connection)
09:51:58 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
09:52:09 Join Ribs [0] (n=ribs@
09:52:41linuxstbanotherbrick: Installing the bootloader is now easier than installing Rockbox - just download ipodpatcher.exe and double-click on it.
09:53:15*safetydan wonders if the latest round of Speex SVN changes are worth integrating
09:53:43safetydanlooks like there's some memory use reductions and fixed-point fixes... so probably
09:53:47 Quit mcphail (Read error: 110 (Connection timed out))
09:55:03 Quit GodEater_ ("CGI:IRC (Ping timeout)")
09:55:36JdGordonpondlife: sup?
09:55:47linuxstbsafetydan: Are there many changes in Rockbox speex compared to speex svn?
09:56:03safetydanlinuxstb, none that aren't related to making it compile
09:56:36linuxstbI mean could you offer patches to the offiiclal tree?
09:56:38pondlifeJdGordon: I'm just trying to see if I've borked root_menu
09:57:35pondlifePressing SELECT during playback no longer goes back to the right track. Is this a known issue still?
09:58:04safetydanlinuxstb, you should be able to generate usable patches yes
09:58:24JdGordon? select from where?
09:58:44JdGordonwhere does it put you?
09:58:49pondlifeTrack 01
09:59:05pondlifeCan you look at a diff for me...
09:59:20 Join _Veseliq_ [0] (
09:59:52pondlifeThe bit of logic I removed is the (last_screen != GO_TO_ROOT) bit
10:00:41bluebrothertucoz: ping
10:00:47pondlifeI find that last_screen == GO_TO_ROOT when I first start up, so this prevents me triggering the init, which is silly.
10:01:59pondlifeI have to work now, but I'd be very grateful if you could svn up and play with it.
10:02:22 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
10:02:24pondlifeIt might have already been broken, I don't know yet.
10:02:49 Quit GodEater_ (Client Quit)
10:02:59JdGordonis this problem only in the db or the tree also?
10:03:05pondlifeDB only I think
10:03:09 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
10:03:21pondlifeFollow Playlist = YES, but IIRC that's ignored in the DB
10:03:21GodEater_safetydan: are you Mr. Speex ?
10:03:28 Join Vyrus001 [0] (
10:04:15JdGordonprobably just needs to remmber tc->selected_item
10:04:28pondlifeJdGordon: Can I leave it with you?
10:04:58 Quit Vyrus (Read error: 104 (Connection reset by peer))
10:04:59JdGordonnot a good idea... im 3/4 asleep
10:05:07pondlifeOK, whenever.
10:05:08JdGordonand its only 8pm :p
10:05:13pondlifeHa , 9am here
10:05:21pondlifeI'm late!!
10:08:07safetydanGodEater, I was the one foolish enough to accept the patch and then upgrade it to the latest version yes :)
10:08:20safetydanI've seen your bug about stereo decoding
10:08:36GodEater_ah ok :)
10:08:49GodEater_I won't pester you about it - just wanted to make sure someone was checking flyspray :)
10:10:48GodEater_I'm very impressed with the format - the whole first disc of HitchHiker's radio series came down to about 41M
10:11:18safetydanIt's pretty damn good at compressing speech
10:11:32GodEater_doesn't do a bad job with the incidental music either
10:11:39 Join pixelma [0] (i=pixelma@
10:13:22safetydanoh goody, there's warnings when I compile with the new speex
10:18:51MeKi||ais there a way I can fix the misaligned text for the themes I dl?
10:19:04bluebrothermisaligned text?
10:19:12bluebrotherdid you install the fonts package?
10:19:43MeKi||aer, whoops.
10:19:53MeKi||aI think I missed that.
10:19:54MeKi||athanks hehe
10:19:56bluebrotherit's explained in the manual ...
10:20:22 Quit anotherbrick ("—I-n-v-i-s-i-o-n— 2.0 Build 3515")
10:21:33MeKi||aoh somehow it uninitiates itself too now
10:21:51MeKi||aits probably something stupid on my part, but
10:21:53 Quit linuxstb ("CGI:IRC")
10:22:20MeKi||ayeah cause when I do 'initiate now', I can see the ipod.control folder
10:22:32MeKi||aand then I go in there manually to play whichever mp3 (probably did that wrong too)
10:22:36bluebrotheryou mean "initialize not
10:23:00MeKi||asorry, lack of sleep.
10:23:22bluebrotheryou can't see the ipod_control folder but in the file view
10:23:42bluebrotherand for that folder to show you need to set the "show files" to "all"
10:23:53pondlifeMeKi||a: You may want to get a new build, I broke the Database option slightly so it would prompt for a new initialisation when it didn't need to, but should be fixed as of 30 mins ago.
10:24:13bluebrotheroh, it was broken?
10:24:18pondlifeA bit.
10:24:32bluebrotherI rarely use the db :)
10:24:40pondlifeI thought that the tagcache initialized status referred to the db, not the thread
10:24:50MeKi||aahh sweet thanks
10:24:54pondlifeIt might still be broken, I'm a database n00b too.
10:25:55JdGordonpondlife: i dont get the code.... selected_item doesnt seem to be updated from the gui list?
10:26:39pondlifeJdGordon: I don't know. I didn't intend to change anything, just inserted a block to do the database init/status.
10:26:57JdGordonI dont think the db ever stored its position properly.. so dw aobut it
10:27:11pondlifeAs long as I didn't knack it
10:27:31pondlifeI thought it used to at least remember the last selected item though.
10:28:06JdGordonit remembers the last track played here... but if you go back and change selection it doesnt remember
10:28:48linuxstb_bluebrother: Do you know about references in LaTeX? I'm trying to add a reference to section 2.2.2 in the ipod bootloader instructions, but the link is just coming up as "section ?? (page ??)"
10:29:55linuxstb_I've looked at how the database is linked, and the section is written as "\section{\label{ref:database}Database}", but for the installation instructions, it's written "\section{Installing Rockbox}\label{sec:installing_rockbox}"
10:30:51 Join webguest34 [0] (i=c023110a@gateway/web/cgi-irc/
10:30:55bluebrotherboth work.
10:30:59pondlifeHmm, the wiki is now hosting Cepstral voice files. Someone should read this:
10:31:11bluebrotherI'm a bit unsure which way I should consider better.
10:32:48linuxstb_OK, my problem was I was linking with ref:installing_rockbox, not sec:installing_rockbox...
10:33:22bluebrotherhehe ... the naming is unfortunately a bit inconsistent.
10:34:28linuxstb_OK, committed.
10:35:27peturpondlife: we should better remove those file then...
10:35:53petur$200 for 1 voice
10:35:56*linuxstb_ forgot that he bid on a Gigabeat and discovers he won the auction...
10:36:15peturhow much?
10:36:16pondlifepetur: Also the Simon voice might well be from
10:36:22linuxstb_55 UKP for an F20
10:36:43pondlifeAlthough I can't see an overall legal status for their normal TTS
10:36:52linuxstb_Expensive in terms of US prices I think, but cheap in the UK.
10:36:54 Join rp- [0] (n=rp@
10:37:29pondlifepetur: Who's legally minded here? You?
10:38:30peturnope :)
10:39:20tucozbluebrother, pong
10:39:21MeKi||ayay works perfect
10:39:23MeKi||athanks guys =D
10:42:02 Quit MeKi||a ("reboot")
10:42:17 Quit aliask ("lalalaladrunkonaweeknight")
10:42:28bluebrothertucoz: how did you convert the svgs? I did it for the m5 image, but the png doesn't look too good
10:42:41tucozI did it in inkscape
10:42:41bluebrotherespecially the transparency could be improved
10:43:09tucozI select the entire svg, including text and choose export from the filemenu
10:43:18tucozexport or save as
10:43:23tucozi do not remember
10:43:43bluebrotherI managed that. Saved as eps, used epstopdf to create the pdf.
10:43:59bluebrotherbut how did you create the png?
10:44:06pondlifeSlasheri: around?
10:44:23tucozbut you do the same for the png, except that you do export as bitmap or whatever the menu item is called
10:44:28*bluebrother thinks this information could go into ManualHowto
10:44:39bluebrotherah, I used ImageMagick to convert eps to png
10:44:46tucozand use the standard settings
10:45:06*pixelma takes notes :)
10:45:12tucozaha. no, svg->png is the easiest part. That is done in inkscape
10:46:06tucozyes, i agree. as it isn't automated, we should probably have it in ManualHowto
10:46:37tucozpixelma, got your memo and reported back :)
10:47:02 Join hannesd_ [0] (
10:47:25 Join decayedcell [0] (i=3ba79d47@gateway/web/cgi-irc/
10:47:26*tucoz wonders how to make hydrairc highlight when someone talks to me
10:48:57*bluebrother just added the steps to LatexGuidelines
10:49:17bluebrothermaybe I can add an additional Makefile rule that can do this automatic
10:49:43bluebrotherso when needed one can simply run that rule manually.
10:49:43tucozthat forces people to have inkscape installed to build the manual then
10:49:51 Join Arathis [0] (
10:49:54bluebrotherbut I need to check if it's possible at all
10:49:58Slasheripondlife: you could just fix the initialized-variable to do the same than the readyvalid
10:50:09tucoztucoz: just checking if highlighting works
10:50:10bluebrothernot if that rule is only used manually :)
10:50:18pixelmatucoz wouldn't have helped if you read it earlier... atm there are problems with connecting to my webspace
10:50:39tucozSlasheri, did you get my message?
10:50:56Slasheritucoz: yes, sounds great
10:51:22Slasheritucoz: i will think about that later, now need to go again ->
10:51:28tucozok. see you
10:52:25pixelmayou can find their wiki in the "help" menu ;)
10:52:53tucozok. thanks
10:53:50pondlifeSlasheri: missed you?!
10:54:22pondlifeOK, well maybe someone else will know.
10:54:41pondlifeI had the database set not to load to RAM and it was working, albeit slowly.
10:55:01pondlifeI then set it to load to RAM and rebooted, but now it needs reinitialising.
10:55:07pondlifeIs that normal/correct?
10:55:34decayedcellpondlife does select still work for you in Database mode, as in go back to your list of albums/artists?
10:55:44pondlifeKind of.
10:55:52*tucoz checks highlighting
10:55:58pondlifeI'm looking into it
10:56:06*bluebrother hilights tucoz
10:56:13pondlifetucoz: Hi
10:56:18tucozthanks guys
10:56:21decayedcellpondlife I don't have that problem with load to RAM on the iPod, but the select button seems to be playing up
10:56:22tucozit works :)
10:56:36pondlifedecayedcell: What happens for you?
10:56:54pondlifeAnd when did it break :/
10:57:11decayedcellpondlife: Say I play a song, and it goes to the WPS. In the WPS screen, if I press select, it opens up file mode and shows me a list of files on my iPod
10:57:42pondlifeHmm, I don't think I changed that. Will try a local revert though
10:57:45decayedcellI think it was broken somewhere after the startup brightness glitch was fixed
10:58:17pondlifeProbably was me then.
10:58:28 Join MeKi||a [0] (
10:58:36pondlifeAnyway, I am aware. Might not get time to look at it until the weekend though
10:58:49LinusNlazy bastard!
10:58:58pondlifeHah, my boss is going to kill me anyway
10:59:16*pondlife shouldn't even have IRC open
10:59:32petursame here :)
10:59:44*pondlife fears that LinusN will be hunting him down now
11:00:14LinusNpondlife: i have sent chuck norris to hunt you down
11:00:45linuxstb_He must be about 70 years old now...
11:01:10*linuxstb_ googles
11:01:14pondlifeI took his glasses off, and he fell down a ravine.
11:01:32linuxstb_Yes, born 10 March 1940...
11:01:36pondlifeA deep ravine full of red build tables .
11:03:04 Quit hannesd__ (Read error: 110 (Connection timed out))
11:08:36 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
11:09:10***Saving seen data "./dancer.seen"
11:09:27*amiconn looks at LinusN wonders what's the status of H300 USBOTG tracing
11:09:48LinusNah, that one
11:10:06LinusNwhat was missing?
11:10:13GodEater_LinusN: you should publish your todo list
11:10:19amiconnpetur wanted some connections traced
11:10:27GodEater_in discrete volumes if necessary
11:10:29 Quit petur ("CGI:IRC (EOF)")
11:10:42 Join p3tur [0] (i=d4efd6a6@gateway/web/cgi-irc/
11:11:04amiconnI would be interested in someting related: if and where the various status outputs of the cypress bridge are connected in the H300 (and ideally, H1x0 as well)
11:11:07 Nick p3tur is now known as petur (i=d4efd6a6@gateway/web/cgi-irc/
11:11:38LinusNamiconn: i tried to trace them, but some of them seem unconnected
11:12:50peturI just wanted to know for sure where ID and VBUS go to as I don't get further then waiting for vbus change when connecting to a device
11:12:52amiconnThat's very possible
11:14:04LinusNpetur: so PortPinAssignments is wrong about ID?
11:14:39peturI don't know, but something isn't working the way I thought. Could be me though...
11:15:19pondlifeLinusN: Would it be ok if I temporarily reverted my root menu mods using an #if 0 block? Hopefully fix it up at the weekend.
11:15:28LinusNpondlife: sure
11:15:37pondlifeOK, I'll do that and stop worrying
11:17:24LinusNpetur: how is vbus supposed to be sensed? via the 1362?
11:18:26peturI think so. But I'm not sure I'm even getting as far as switching it on. So if it's generated by internal or external chargepump...
11:18:41peturI really should try to measure
11:20:13 Join inversions [0] (
11:23:44markunpetur: did you see that austriancoder wants to work on a USB stack?
11:24:34peturmarkun: yes, he mailed me. he can have whatever I already have but I don't have the time to do mentor stuff
11:25:58tucozpetur, not even for the good sake that rockbox is? ;)
11:26:42pondlifedecayedcell: Try a new build now.
11:27:03pondlifeAnd amiconn even got some of his old bytes back.. for now.
11:30:33 Join Wiwie [0] (
11:30:46 Quit Wiwie (Client Quit)
11:46:25 Quit jaebird (Read error: 110 (Connection timed out))
11:46:54safetydanlinuxstb_, actually the speex files aren't totally clean. There's a couple of warning fixes in there as well
11:48:56linuxstb_safetydan: What do you think about asking the speex maintainers to commit your changes? Could they be done in a way acceptable to them?
11:49:22linuxstb_It seems to be a codec undergoing a lot of development.
11:49:52 Quit Arathis ("Bye, bye")
11:50:39 Join juxtap [0] (
11:51:00safetydanThe warnings are all pretty trivial so I wasn't sure if it'd be worth the effort given that it's still in development.
11:52:16pondlifeWhy does the playback context menu have the EQ in it? Seems odd to have one sound setting in there when the whole Sound Settings is available anyway.
11:53:51safetydanpondlife, err... at the time people were excited about the EQ and I was swayed by their arguments to include quick access to the settings
11:54:36pondlifeHmm, I'd rather quick access to Vol/Bass/Treble if anything. Should be removed next time anyone's poking around in there IMHO.
11:54:37safetydanat least that's how I think it went anyway
11:55:38JdGordondoes anyone use torrentflux?
11:56:13pondlifeWooh, the "Bass" EQ preset makes a horrible high pitched squeal...
11:57:06safetydanit shouldn't do that
11:57:14safetydanusually only happens if the settings are read incorrectly
11:58:09 Join GreyFoux [0] (
11:58:13 Quit GreyFoux (Client Quit)
11:58:18pondlifesafetydan: They look ok
11:59:10safetydanpondlife, as in the .cfg file looks okay, or the settings look okay in the menu?
11:59:11pondlifeOK, I can repro with only one band enabled.
11:59:31pondlifePK2: 0.5dB, 800Hz, 1.0Q
11:59:39pondlifeAccording to the menu
12:00:21safetydanWhat happens if you set that manually?
12:00:55preglow0.5 db?
12:00:56pondlifeRebooting fixes
12:01:00preglowit should be 0.7
12:01:06preglowor no, 1.0db for peak filters
12:01:20preglowsans db...
12:01:30safetydanlinuxstb_, I just subscribed to the speex dev list so we'll see what happens
12:01:32preglowi'm confused, forget that
12:01:50preglowpondlife: what dap is this?
12:02:20preglowi remember jdgordon messed up one of the defaults
12:02:27preglowi think that was for lowshelf
12:02:39safetydanpreglow, that was a while ago wasn't it?
12:02:39preglowJdGordon: when you did the new settings system
12:02:47JdGordonwas it fixed?
12:02:53preglowsafetydan: well, yes, but it'll still stick around if not used
12:03:08pondlifeHaha, there's another browser bug too
12:03:17safetydanpreglow, true, but I think pondlife was loading that from a preset
12:03:28pondlifeI selected "acoustic" once, now it auto selects every time I go in!
12:03:41preglowwell, how practical
12:03:51preglowcan you already now see how rockbox saves you from all the extra work?
12:03:56preglowit knows what you wants anyway
12:04:07pondlifeBut it's guessing wrong
12:04:28pondlifeAnyway, I can't reproduce the squealing, thank monster
12:07:45 Join anathema [0] (
12:08:39 Join kubiix [0] (
12:11:44 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
12:12:09 Join miepchen^schlaf [0] (
12:13:45 Join omin [0] (
12:20:45 Quit _fejfighter_ ()
12:24:13GodEater_is there a tag editor which works with speex files ?
12:24:37safetydanGodEater, well they're just Ogg stream's so anything that works for those should work for speex
12:24:58GodEater_safetydan: thought so - just wanted to check :)
12:25:30 Quit inversions (Read error: 110 (Connection timed out))
12:29:31safetydanGodEater, well upgrading the libspeex hasn't fixed the stereo files issue
12:29:42safetydanseems to be a bit of a performance boost though
12:32:13GodEater_hmm, easytag doesn't seem to understand the speex "title" and "author" tags - it's not displaying them
12:32:57 Quit GodEater_ ("CGI:IRC")
12:33:10 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
12:35:25 Quit juxtap (Read error: 110 (Connection timed out))
12:36:45 Quit linuxstb_ ("Leaving")
12:37:33 Quit idnar (Nick collision from services.)
12:37:35 Join idnar_ [0] (n=mithrand@unaffiliated/idnar)
12:40:29preglowsafetydan: gonna commit that sync?
12:40:48safetydanpreglow, probably not tonight. Want to track down the stereo issue first.
12:40:57preglowsafetydan: what i'm wondering about more than the stereo file issue is the bloody burst issue
12:41:08safetydanburst issue?
12:41:50preglowi get (got) noise bursts from time to tiem
12:42:08safetydanah that... what encoder version did you use?
12:42:45GodEater_preglow: Llorean got those I think, he was using some ancient encoder
12:42:57GodEater_I'm not getting any with the 1.1.12 encoder I'm using
12:43:41GodEater_hah - resampled to 32Khz, dropped to mono, encoded the entire CD to speex and the total size is 14MB!!!!
12:44:20markunGodEater_: which other settings did you use?
12:44:54safetydanGoing to have to leave it there. Night all.
12:44:54 Quit safetydan ("Ex-Chat")
12:45:10GodEater_markun, none, that was it
12:45:16 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
12:45:29GodEater_no special options issued to speexenc at all
12:45:49GodEater_and it sounds great
12:46:41preglowi didn't use an encoder, i just found some files
12:47:00preglowbut what has that got to do with anything? i sincerely doubt that whatever encoder was used, it did not produce noise bursts
12:47:17preglowjust rephrase that to make sense and you'll get my meaning
12:47:43GodEater_I get it - I'm just telling you what Llorean found. Although he did say foobar played them back fine from the old encoder =/
12:47:53GodEater_he only got the noice in RB too
12:49:07markunGodEater_: last time I tried it I couldn't make speex sound better than ogg vorbis with the same size
12:49:12markunI'll try again
12:49:59preglowtalking pure speecch?
12:50:28markunsynthesized speech
12:50:31 Join PaulJam [0] (
12:50:33markunperhaps that's different
12:50:56preglowin a way it should be ideally suited
12:51:06preglowbut depends on tuning
12:51:32markunbetter would be to just have a espeak plugin in rockbox, a much better compression rate :)
12:51:42preglowah, espeak
12:51:46preglowthat'd a bit different, yes
12:52:03markunHis esperanto is quite good :)
12:52:32GodEater_markun: did you try dropping the sample rate and from stereo to mono as I did ?
12:52:49preglowthere's no reason to keep speech as stereo
12:52:51markunI made some improvements to the 'r' in Dutch which I think will improve German and French (and possibly other lanugages)
12:53:01preglownot unless you're teleconferencing with several people or something
12:53:05markunespeak's output is 22050Hz mono
12:53:23markunI could lower it to 16kHz maybe
12:54:00preglowmarkun: i'm almost certain speex will compress natural speech better than espeak
12:54:10preglowuse 16kHz
12:54:24GodEater_speex is optimised for 8,16 and 32 Khz
12:54:33preglow16khz is recommended
12:54:37markunI'll give it a go with a audiobook
12:54:41preglowi don't think i'd consider using 32 khz
12:55:01GodEater_seems fine to me at 32Khz
12:55:15 Join kubiix [0] (
12:55:24preglow32khz is a bit overkill for voice, is all
12:55:45GodEater_true - but there's incidental music in what I'm listening to
12:55:59markundon't know how well transcoding from mp3 or ogg will affect the quality of the speex file..
12:56:03preglowthen i'd consider using something else than speex
12:56:14GodEater_it's *very* incidental
12:56:17preglowmarkun: not positively, that's for sure
12:56:18GodEater_a few seconds at a time
12:56:31markunpreglow: no, that's what I thought..
12:56:32 Nick idnar_ is now known as idnar (n=mithrand@unaffiliated/idnar)
12:56:44preglowi think i've got an interview on some cd around here
12:56:47preglowwill try to encode that
12:56:51markunbut I don't have any other speech sources
12:57:00preglowseems i have speexenc 1.1.12
12:57:03preglowis that any good?
12:57:31markunpreglow: ok, can you also encode with ogg vorbis at the same nominal bitrate?
12:58:37preglowi'll try
12:58:50preglowjust need to free some space on my tightly packed linux partitions for ripping :>
12:59:29 Join barrywardell [0] (
12:59:30preglowmarkun: don't think i can bothered to use any other oggenc than what's default in ubuntu, though
12:59:46preglowmarkun: should i resample to 16khz for vorbis too?
13:00:11markunyes, that seams fair
13:00:22preglowi don't think vorbis is very well tuned for that, but we'll see
13:01:03markunyou could try a non resampled version for vorbis too at the same bitrate
13:01:45pondlifeJdGordon: I've found a recipe for the root menu database selection problem. Shall I Flyspray it?
13:01:57JdGordonor juyst commit it
13:02:08pondlifeNo idea what's causing it :(
13:02:14preglowJdGordon: recipe to avoid 64 bit problems: use intptr_t instead of void *
13:02:15LinusNwow, espeak really takes some getting used to
13:02:24preglowLinusN: it's not very high quality, really
13:02:37preglowit uses some weird synthesis method
13:02:40bluebrotheris there a recent espeak voice file around somewhere?
13:02:51JdGordonpondlife: oh, misread... yeah fs it
13:03:31bluebrotherfunny wording ... to flyspray something :)
13:03:40markunbluebrother: I could try to make you one
13:04:07bluebrotherI failed to compile espeak on this machine. Haven't tried on my other
13:04:11markunfor which target?
13:04:14bluebrothermarkun: h120
13:04:38JdGordonpreglow: yeah, ive been in a daze all day and shouldnt have commited this arvo... amiconn to the resuce tho so all is fine now :p
13:05:02LinusNi tried to create a festival voice file a few days ago, and every phrase ended with noise burst :-(
13:05:08 Quit JdGordon ("bed time")
13:05:13preglowJdGordon: at least use that whenever you know you're going to cast it to/from an int at some time
13:05:24preglowLinusN: picky, aren't you
13:05:39 Join FOAD_ [0] (
13:05:59pondlifeThat noise burst was the full stop perhaps?
13:06:20LinusNi have no idea
13:06:31pondlifeForgot to smile
13:06:38LinusNneither the wav files or the mp3 files have it
13:06:39 Quit omin ()
13:07:02LinusNbut the compiled voice file has it
13:08:27 Quit decayedcell ("CGI:IRC (EOF)")
13:08:38markunLinusN: do I need to compile rockbox to build a voice file or am I doing something wrong?
13:08:39preglowdamn, this rip was slow
13:09:11LinusNmarkun: ../tools/configure - then select a platform then (V)oice
13:09:13***Saving seen data "./dancer.seen"
13:09:13preglowRIGHT, it did the whole album
13:09:22LinusNmarkun: then you type "make voice"
13:09:39LinusNmarkun: you must have a speech synth and lame in the path
13:10:09markunvoice: not found
13:10:09markunencode: not found
13:10:27 Quit anathema (Read error: 113 (No route to host))
13:10:31markunhm, I think I know what's wrong
13:10:33bluebrotherthose should be in the tools folder
13:10:47bluebrotherah, no, it was makevoice
13:10:52bluebrotheror something like that
13:12:30pondlifehaha, database progress: 120%
13:13:17pondlifeShould it be marked as Ready when it's loading into RAM still? I guess it works from disk in the meantime...
13:13:21bluebrotherhmpf, I shouldn't do three things at once
13:13:28preglowsomeone really needs to bloody make a _GOOD_ open source sound editor
13:13:54Shaiddon't like Audacity?
13:13:55GodEater_audacity sucks too much huh ?
13:14:01preglowoh sweet lord, yes :/
13:14:27Shaidtried sweep?
13:14:41preglowyes, that just crashed and corrupted my rip
13:14:43preglownow i need to rip again
13:15:20preglowi don't get sweep, they brag and brag about the scrubber functionality, but the sound is horrible
13:15:26preglowit doesn't even try to interpolate
13:15:26Shaidyou're having lots of luck tonight
13:15:42preglowi asked them about it, and they said making it interpolate wasn't trivial
13:15:47preglowat which point i stopped counting on them having a clue
13:17:12ShaidI guess there's always snd if you're feeling hardcore
13:17:21LinusNpreglow: "someone really needs to bloody make a _GOOD_ open source sound editor"
13:17:31LinusNpreglow <- someone
13:17:45preglowLinusN: i've thought about it more than once, but time...
13:17:51preglowit's quite a lot of work
13:17:56LinusNah, that small detail
13:18:32preglowplus, i'd probably be sued by adobe for cloning what adobe audition used to be like before they force-fed it and made it the bloat it is
13:18:46Shaidyou mean cool edit pro?
13:18:49preglowcool edit pro 1.5 is still what i prefer, but that's for windows
13:18:56Shaiduse wine?
13:18:59preglowamd64 :/
13:19:11Shaiduse wine in vmware?
13:19:20Shaidor even windows in vmware.
13:19:30ShaidI think my brain skipped a beat when I said wine in vmware. :D
13:19:38preglowcan't be bothered with vmware
13:19:40preglowtried once, failed
13:19:45 Join austriancoder [0] (n=austrian@
13:20:06peturtry virtualbox, it's quite nice...
13:20:45 Quit FOAD (Read error: 110 (Connection timed out))
13:22:55preglowmarkun: what bitrates/q params?
13:25:06*amiconn wonders what's wrong with audacity
13:25:09 Quit Shaid (Read error: 104 (Connection reset by peer))
13:25:29preglowit's cumbersome, buggy, and doesn't do nearly all i want
13:25:35preglowand has a really sucky spectrum view
13:25:38preglowisn't even realtime
13:26:36 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
13:30:39bluebrotherthe license of cool edit (before it became "pro") was quite funny
13:33:15 Join Febs [0] (
13:34:42 Quit barrywardell (Read error: 110 (Connection timed out))
13:36:03preglowmarkun: vorbis indeed does the job very well
13:37:41preglowbluebrother: oh?
13:37:42 Join Thundercloud [0] (n=thunderc@
13:46:47 Join Nico_P [0] (
13:49:04 Join jaebird [0] (
13:49:46XavierGramiconn: does the ondio take power from the USB port?
13:50:01 Quit GodEater_ ("CGI:IRC (Ping timeout)")
13:50:52 Join GreyFoux [0] (
13:50:54 Quit GreyFoux (Remote closed the connection)
13:51:27XavierGrnevermind I found out myself
13:51:40XavierGreven without batteries it will power on and connect to my PC
13:52:32XavierGrwell the sad part is that I did a batter_benchmark and with those crappy batteries in the box it managed to playback for only 1 hour and 10 minutes :(
13:52:48XavierGrI will try again with good batteries but I am afraid that in the end my Ondio suffers from the power drain issue too.
13:53:27XavierGrI don't mind to do a surgery on it, but I wonder where I will find the chip that needs to be replaced
13:54:06 Join Arathis [0] (
13:56:08pixelmaXavierGr: oh? I thought you said it didn't get warm...
13:57:27pixelmaalso: be a bit carefull - amiconn suspects that connecting the Ondio without batteries in it could cause this issue - but I don't know
13:58:32 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
13:58:55preglowmarkun: btw, did test the new crossfeed filteR?
13:58:57preglowdid you, that is
13:59:04XavierGrpixelma: it didn't get warm yesterday that I was holding it, but then again only 1 hour?
13:59:05 Join inversions [0] (
13:59:21XavierGrand the USB trick without batteries was done after the bb so it shouldn't have affected it
13:59:39 Join anathema [0] (
13:59:43XavierGrI will buy a new AAA alkaline set to do another test too
14:00:03pixelmano - I meant it could do the damage to the power regulation chip (or whatever)
14:00:26XavierGrthough voltage droped from 4,57 to 2,28 in just 1 hour and 10 minutes
14:00:37 Join merlin2049er [0] (
14:00:50XavierGrpixelma yes, got that
14:02:02 Join Stalwart^ [0] (
14:06:54XavierGrcorrection: it dropped to 2,82 not 2,28
14:07:06pixelmajust wanted to say that it shouldn't get worse... or in case it really was only caused by weak batteries...
14:16:03 Quit idnar (Read error: 145 (Connection timed out))
14:17:06 Quit PaulJam (".")
14:18:14 Quit linuxipod ("Chatzilla 0.9.77 [Firefox]")
14:18:22 Quit Stalwart (Read error: 110 (Connection timed out))
14:18:44 Join idnar [0] (i=mithrand@unaffiliated/idnar)
14:18:53 Quit inversions (Read error: 113 (No route to host))
14:21:11 Quit Febs (Read error: 110 (Connection timed out))
14:23:40 Quit merlin2049er (Client Quit)
14:25:01 Nick Stalwart^ is now known as Stalwart (
14:25:57 Join pearldiver [0] (
14:29:49 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
14:30:09 Join miepchen^schlaf [0] (
14:30:11markunpreglow: crossfeed sounds fine to me
14:30:25 Quit Ribs ("eh eh ehhhh!")
14:30:42markunwas speex slightly better than speex or wasn't there much diffrence?
14:30:56markunspeex better than vorbis :)
14:33:37preglowi asked you what bitrates you wanted
14:34:05preglowat 44.1khz i can't really push the vorbis one too low
14:34:15preglownot compared to the speex one, at least
14:34:58markunthe same bitrate with the same audio file would be the best test I think
14:35:21markunbut I understand that vorbis might not be optimised for those framerates
14:35:25markunwhat did you test?
14:35:56 Join Rondom [0] (
14:45:54 Quit webguest34 ("CGI:IRC (EOF)")
14:46:13 Quit SirFunk (Read error: 110 (Connection timed out))
14:46:36 Quit GodEater_ ("CGI:IRC (Ping timeout)")
14:59:37 Quit Arathis (Remote closed the connection)
15:00:36 Join Arathis [0] (
15:01:31 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
15:09:01preglowwill upload some files in five mins
15:09:16***Saving seen data "./dancer.seen"
15:12:15 Join OgMaciel [0] (n=omaciel@ubuntu/member/gnukemist)
15:14:23markunpreglow: maybe we can get something from LibriVox in a lossless format
15:15:03 Join datachild` [0] (
15:15:15jhMikeShmmm...two things cure all the ata driver ills: 1) spin locking the mutex 2) guaranteeing the waiting thread runs once per scheduler round in not too long
15:16:09markunpreglow: thansk
15:19:31 Join webguest72 [0] (i=436769f2@gateway/web/cgi-irc/
15:20:10 Quit idnar (Nick collision from services.)
15:20:11 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
15:21:27preglowmarkun: and that's lossless, flac should be there as well
15:21:29 Join SirFunk [0] (
15:21:44markunpreglow: yes, the speex versions sound a lot better
15:22:04 Part tucoz
15:22:44 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
15:23:03markunpreglow: speex encoding will not be easy to do with rockbox, right?
15:23:12preglowmarkun: might not be impossible, wouldn't know
15:23:13markun(real time)
15:23:40markunfox recording lectures or something it would be great
15:23:56preglowbeen thinking about it myself
15:24:06preglowbut i wanna optimise decoding first, maybe i'll have a better understanding then
15:24:44 Quit datachild (Read error: 110 (Connection timed out))
15:24:55 Quit webguest72 ("CGI:IRC")
15:25:47preglowmarkun: i used vbr, though, but i guess there's absolutely no reason not to do that
15:25:51preglowit's not like i'm streaming or anything
15:26:34markunare you not talking about cbr?
15:26:59preglowthe one marked cbr is cbr, the other is with −−vbr
15:27:01markunAh, no readon NOT :)
15:27:08markunthis connection is driving me crazy..
15:27:12markunI can't see what I type
15:27:19preglowknow what that feels like, yes
15:27:20markunwell, not in time
15:27:40preglowthe vbr speex one sounds really nice, i think
15:27:54markunyes, very
15:28:12preglowquality 5 too
15:28:30preglowah, no no, quality 7
15:28:34 Quit blueworm ("Leaving")
15:29:17preglowquality 5 sounds nice, though
15:29:22preglownot as nice, but nice
15:29:28jhMikeSI should mention too in case it's important enough that disk access times go way down in general without thread blocking
15:30:06preglowjhMikeS: good to know, the new scheduler is still slightly controversial :)
15:30:52 Quit XavierGr ()
15:30:57jhMikeSHigh speed locking with spins is much better. I made it so a mutex can be lock in either way.
15:31:49preglowmarkun: i think quality 5 still sounds better than the vorbis one
15:32:12preglow265k vs 320 for vorbis
15:32:26 Join midkay [0] (n=midkay@rockbox/developer/midkay)
15:32:38preglowspeex is ideally easier to decode too
15:32:42preglowcurrent code needs optimising
15:33:09jhMikeSDidn't know it was still controversial though...needs tweaks and more refined use of features otherwise it's quite nice to have.
15:34:07amiconnpondlife: What kind of navigation problems did the db autoinit cause?
15:34:35 Quit petur ("worrrk")
15:35:06pondlifeamiconn: It caused some unexpected browser locations when SELECT was pressed from the WPS
15:35:29pondlifeHowever I have since managed to repro similar problems with that reverted.
15:35:58pondlifeI keep getting a lockup with the disk in use, very similar to the problem jhMikeS is looking into.
15:36:32pondlifeProbably unrelated but a PITA all the same.
15:37:53 Join XavierGr [0] (
15:39:54pondlifejhMikeS: Are you getting near a fix yet?
15:39:54markunamiconn, preglow: don't you want to help mentoring for GSoC?
15:40:02jhMikeSthe higher the priority the thread that tried to use the drive relative to another using it, the more it was blocked. :\
15:40:25preglowmarkun: i don't know if i'll have time
15:40:44jhMikeSpondlife: just have to make sure having to locking modes for a mutex unlocks in the best way
15:41:13 Join Criamos [0] (
15:41:21markunI'm not sure either how much time I will have exactly
15:41:24jhMikeSbut a recording flush is almost instant with database building going on at the same time
15:41:27 Quit GodEater_ ("CGI:IRC (Ping timeout)")
15:41:41pondlifejhMikeS: OK, I think I understood that. It means you commit in about 10 mins, right ;)
15:42:12jhMikeSnot really....this has to be very right before doing that
15:42:35pondlifeWell, it's pretty wrong at the moment
15:42:58preglowmarkun: only project i can see that i'd feel comfortable in mentoring would be the wma codec one anyway
15:43:17pondlifeI'm finding that I get a very simlar lockup during playlist saving when using database.
15:43:26jhMikeSand it needs the atomic version for dual core
15:43:45jhMikeSprobably the same situation
15:44:01preglowonly dual core we have is arm
15:44:14preglowand the only facility ours has for atomic operations is the swp instruction
15:44:26pondlifeI left it for 40 mins but it never came back to life.
15:45:17jhMikeSI think that should be just the thing...I can use test_and_swap. It'll return false again if something else set the variable right?
15:45:41jhMikeSit uses swpb
15:45:49preglowwhere's test_and_swap?
15:45:49 Quit XavierGr ()
15:46:09 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
15:46:15jhMikeStest_and_set rather :)
15:46:15preglowi sure can't find it
15:46:37jhMikeSline 662
15:46:48preglowi do know how to grep ;)
15:47:44preglowshould be good
15:48:08*pondlife is getting annoyed with a barking dog up the street
15:48:31preglowi'm annoyed by dogs in general
15:48:36jhMikeScool...I can't figure why we need two function implementations though
15:49:01preglowyou mean the non-atomic version as well?
15:49:26 Join juxtap [0] (
15:49:29preglowwell, code up test_and_set for coldfire and sh as well, and we're ok
15:49:40 Join funky [0] (n=repulse@unaffiliated/funky)
15:49:45 Join XavierGr [0] (
15:50:32 Quit BiptoN ("CGI:IRC (EOF)")
15:50:38markunpreglow: and for me it would be espeak I think
15:50:55pregloware you sure espeak is suitable?
15:50:59preglowi don't really know what algos it uses
15:51:01jhMikeSsure, but it can just be a non-atomic version or whatever
15:51:02preglowbut it doesn't sound like lpc
15:51:12preglowjhMikeS: could, sure, but would be nice to have it done
15:51:15jhMikeSprobably doesn't even needs asm
15:51:18preglowi don't really care either way, really
15:51:46markunpreglow: it was first designed for a 12MHz ARM (acorn) and the guy who is writing it is really helpful
15:52:02preglowdone any research on other stuff anyway, though?
15:52:14preglowespeak does take some getting used to
15:52:30markunI checked flite, but I don't think we want that
15:52:37jhMikeSare we talking "voice synth" here?
15:53:22jhMikeScool...I've been wondering if we could have that for awhile instead
15:53:45markunpreglow: maybe some rockbox users become interested in helping improve the sound of espeak when we have a plugin for it
15:54:25markunjhMikeS: it will not sound as good as offline created voice files, but much more flexible
15:54:37jhMikeSso voice files would be phonetic text instead of audio clips?
15:55:31markunperhaps we should first try to make it into a plugin and then later decide if we want to integrate it into rockbox
15:56:02jhMikeSif it can sound like a Speak 'n' Spell I'd be just giddy
15:56:13preglowdamn, that'd be cool
15:56:18preglowspeak & spell is pure lpc
15:56:21preglowshould be easy
15:56:48preglowlinear prediction coding
15:56:52 Join desowin [0] (
15:56:52preglowbasically just a big iir filter
15:56:59preglowflac uses it, speex uses it
15:57:04preglowalthough not for making nasty vocal samples :>
15:58:29 Quit rp- ("leaving")
15:58:49jhMikeSnot so nasty to me...:)
16:00:33jhMikeSI'm of couse talking about getting every indecipherable syllable absolutely genuine sounding
16:01:03markunpreglow: btw, don't be fooled by the fact that eSpeak is C++, it's almost all just normal C
16:01:19preglowmarkun: don't view that as much of a problem anyway, i don't think it's much code
16:04:45 Quit darkless (Remote closed the connection)
16:05:28markunIt's starting to worry me that I haven't received any SPAM in my gmail account for a day.. :)
16:08:20preglowi've got a can of it in my bookshelf if you want it
16:10:36GodEater_markun: I run a greasemonkey script that hides the count of spam from me anyway
16:11:14 Nick datachild` is now known as datachild (
16:13:08*GodEater_ has now been hunting all day, and hasn't found a single linux program which can read/write tags to speex files
16:14:40markunpreglow: nah, thanks :)
16:16:07jhMikeShmmm...might do as amiconn suggested and make mutexes tighter with the scheduler in some ways. could keep it cleaner.
16:18:41 Join kubiix [0] (
16:18:51 Quit kubiix (Client Quit)
16:33:16 Join dmdfan- [0] (
16:33:21 Join CriamosAndy [0] (
16:33:25 Quit ender` (Read error: 104 (Connection reset by peer))
16:34:04 Join ender` [0] (
16:34:07 Quit Criamos (Read error: 104 (Connection reset by peer))
16:35:45 Quit B4gd3r (
16:35:45 Quit qwm (
16:35:45 Quit Kasperle (
16:35:45 Quit preglow (
16:35:49NJoinB4gd3r [0] (
16:35:49NJoinqwm [0] (
16:35:49NJoinKasperle [0] (
16:35:49NJoinpreglow [0] (n=thomjoha@rockbox/developer/preglow)
16:39:10 Join bonbonthejon [0] (
16:40:39 Quit GodEater_ ("CGI:IRC (Ping timeout)")
16:43:28 Join toffe82 [0] (
16:47:55 Join perl|work [0] (
16:49:57 Quit dmdfan (Read error: 110 (Connection timed out))
16:51:00 Quit Rondom ("Ex-Chat")
17:04:42 Join zylche [0] (
17:06:30 Join rm [0] (i=irc@gateway/tor/x-7121913ab26d8195)
17:08:20 Quit austriancoder ("Lost terminal")
17:09:19***Saving seen data "./dancer.seen"
17:11:34 Part rm
17:16:52pondlifeSlasheri: A bit of reproducible database weirdness here!
17:28:02 Join ]MrFWorK[ [0] (n=1@
17:28:13 Part ]MrFWorK[
17:29:05 Quit pearldiver (Read error: 110 (Connection timed out))
17:31:47 Join web-taz [0] (
17:37:25 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
17:44:57 Join rotator [0] (n=e@rockbox/developer/rotator)
17:49:20 Join lowlight [0] (i=c730180a@gateway/web/cgi-irc/
17:50:55 Quit GodEater_ ("CGI:IRC (EOF)")
17:54:02 Join SimonSelki [0] (i=aa4c14fd@gateway/web/cgi-irc/
17:55:22SimonSelkiHey, I'm a little confuzzled on the issue of recording with rockbox on an iPod
17:55:26SimonSelkiis it possible?
17:55:39SimonSelkiI'm on the mailing list, and there hasn't been any talk about it
17:55:48SimonSelkiI can't find anything in the manuals either
17:59:45markunafaik its possible
18:00:03markunBut you need a dock connector for a microphone
18:01:52SimonSelkiHow do you mean
18:02:17SimonSelkilike, I was about to try recording with an audio i/o cable from another iPod
18:03:00markunand it didn't work?
18:03:06SimonSelkinot a very practical way of transfering songs from one DAP to another, but you get what I was trying to do
18:03:21SimonSelkino, I thought I'd consult anyone here before I gave it ago
18:03:32SimonSelkia go that is
18:03:56markunwell, I'm not a ipod expert, so perhaps you should wait for someone else to answer
18:04:56SimonSelkiI have the right idea though? Is that the way the recording was meant to work? I was also going to try recording from a studio, using a mic and a sound board
18:05:57 Join kubiix [0] (
18:07:23 Quit kubiix (Client Quit)
18:07:28SimonSelkifair enough. I'll ask later
18:07:42 Quit SimonSelki ("CGI:IRC (EOF)")
18:09:33pondlifeCan someone who uses the database and has the current build on their device try something quick for me?
18:10:40lowlightmarkun: I'm getting a gb f40 that was "accidentally" formatted and now has no os. Do you know how I go about restoring it?
18:10:57markunyes, very easy
18:11:05 Join kubiix [0] (
18:11:44pondlifemarkun: 1) From a restart, go into Database - Artist - (pick an artist) - <All Tracks> - (pick the first track).
18:12:24pondlife2) On the WPS, press SELECT once
18:12:32pondlifeWhat is displayed?
18:12:42 Quit kubiix (Client Quit)
18:13:13pondlifeI get the file browser (not the database browser). Normally in the root directory.
18:13:20pondlifeBut in the simulator it works fine
18:13:32 Join kubiix [0] (
18:13:47markunpondlife: I get the WPS with paused playback
18:13:54pondlifeAh, not PLAY - SELECT
18:14:04markunI did SELECT :)
18:14:10markunRemember that I have a Gigabeat..
18:14:36pondlifeAh, ok - which ever button goes back to the browser without stopping, if there is one
18:14:39markunWell, POWER brings me back to the file browser
18:14:49pondlifeOK, file - not database, right?
18:14:58markunyes, correct
18:15:19pondlifeWhich is wrong, I reckon. And the sim goes back to the database as I would expect...
18:16:08markunlowlight: you can probably just connect to USB and then unzip GBSYSTEM.ZIP from GigabeatFXPort and copy the bootloader over (FWIMG01.DAT)
18:16:13 Quit FOAD_ ("I'll be back")
18:16:28 Join FOAD [0] (
18:17:21roolkumarkun: is there a reason select and play are swapped on Gigabeat (I have it swapped back in my local build an completely forgotten about it)
18:17:35lowlightmarkun: good to know...thanks
18:19:15markunroolku: I think the iriver buttons wouldn't be like this if the play button wasn't called play..
18:19:55markunbut you think it makes more sense to have POWER pause the music?
18:20:19 Join webtaz [0] (
18:20:42markunright now you can leave the WPS and return to it with the same button, I think that's a good thing.
18:21:14markunAlthough I still think we should make A that button
18:21:20 Join K4rP4D [0] (
18:27:29roolkumarkun: maybe I am used to it from the irivers, but pressing select to start browsing feels natural as you have your finger on the cross already to move up and down
18:27:45perl|workmarkun, but again, i vote for those buttons to be switched
18:27:59perl|worki hope that day will come one day
18:28:19roolkumarkun: and I have A mapped to shut down anyway (to be able to use power as shift in for example the virt keyboard)
18:28:32perl|workmarkun, see? :P
18:31:59 Quit lowlight ("CGI:IRC (Ping timeout)")
18:33:04roolkumarkun: also it is very rare that i return from browsing without having selected a new title which means for me it is the same key to leave and return to the WPS
18:34:37 Join darkless [0] (
18:35:36 Join PanicZone [0] (i=519b0257@gateway/web/cgi-irc/
18:35:58PanicZoneHello Peoples I Have A Problem! :(
18:36:40PanicZoneIf Anyone Is Here!
18:37:08 Quit PanicZone (Client Quit)
18:37:34zei'm sure it was obvious in that minute that nobody is
18:37:49thegeekI love people like this
18:38:11 Quit web-taz (Read error: 110 (Connection timed out))
18:38:12thegeekif they give up this easily imagine what they must be like in real life
18:48:37toffe82perhaps is computer took fire ;)
18:48:44 Join Frode_ [0] (n=Frode@
18:56:26 Join GodEater_ [0] (i=c2cbc95c@gateway/web/cgi-irc/
18:58:44 Join austriancoder [0] (n=austrian@
18:59:17austriancoderhas somebody a dual-boot ready bootloader for sansa to download?
19:00:31austriancodercome on..
19:01:32pondlifeaustriancoder: Good luck tomorrow
19:01:53austriancoderthe train will come in about 3 hours
19:03:26GodEater_ooh, tomorrow is the big day huh ?
19:03:30GodEater_who's going ?
19:05:04austriancoderr- and me
19:05:41 Nick Frode_ is now known as Juice^ (n=Frode@
19:06:15austriancoderbye all
19:06:19 Quit austriancoder ("Kopete 0.12.4 :")
19:09:20***Saving seen data "./dancer.seen"
19:16:37 Quit axion1 (Read error: 131 (Connection reset by peer))
19:16:45 Join axion1 [0] (
19:24:44 Quit JB_Away (Read error: 104 (Connection reset by peer))
19:24:45 Join jborn_ [0] (
19:25:30 Join mirak [0] (
19:27:23 Quit pondlife ("disconnected has pondlife")
19:32:00 Join Ribs [0] (
19:32:43 Join webguest80 [0] (i=598a4ee2@gateway/web/cgi-irc/
19:34:13 Join rp- [0] (i=rp@
19:39:13 Quit webguest80 ("CGI:IRC (Ping timeout)")
19:40:21 Join muesli__ [0] (n=muesli_t@
19:41:17 Quit Ribs (Remote closed the connection)
19:46:25 Join Ribs [0] (
19:52:55 Quit toer ("Changing server")
19:53:07 Join obo [0] (
19:58:28 Quit ender` (Read error: 104 (Connection reset by peer))
19:59:17 Quit Ribs ("eh eh ehhhh!")
20:05:36 Join Febs [0] (n=chatzill@
20:09:40*Nico_P is starting to get results with the wps code rewrite
20:09:56 Quit Hadaka ("leaving")
20:10:16 Quit markun (Read error: 104 (Connection reset by peer))
20:10:22 Join markun [0] (n=markun@rockbox/developer/markun)
20:14:54 Join lee-qid [0] (
20:16:04FebsIt is normal for directory cache to scan the disk on every startup, regardless of whether the disk contents have been changed?
20:20:14 Join blueworm [0] (
20:21:18muesli__how should rbx know if content was changed if not a scan will be performed? :o
20:21:29bluebrotherNico_P: what's the goal of your wps rework?
20:22:22Nico_Pbluebrother: clearner, smaller, more efficient and more maintainable code
20:22:24Febsmuesli__: I just don't remember it taking this long in the past. Each time I turn on the ipod, the scan takes between 1:50 and 2 minutes to complete.
20:22:51Nico_Pbluebrother: it's based on a wps tokeniser project safetydan started
20:23:03bluebrotherah. Other improvements?
20:23:10 Join thomj [0] (
20:23:33Nico_Pbluebrother: do you have some in mind ?
20:23:35muesli__Febs thats kinda takes some secs on my 15gb drive
20:23:59*thomj picks at preglow
20:24:01bluebrotherNico_P: no ... I think the wps has enough features ;-)
20:24:46Kasperleadd an html parser ;)
20:24:48bluebrotherexcept maybe viewports, but that is a different story ...
20:25:14 Quit preglow (Read error: 110 (Connection timed out))
20:25:37bluebrothera simple html parser plugins would be nice. But I'm not sure if that would be feasible at all ...
20:26:00Nico_Pbluebrother: i'm leaving the viewports to amiconn :)
20:26:09 Nick thomj is now known as preglow (
20:26:10preglowabout time
20:26:24Nico_Pbluebrother: a simple html parser would probably be quite doable
20:26:45bluebrothersure. But I'm a bit unsure how much work it would be
20:26:57bluebrotheras it should add support for tables and graphics at least
20:27:04Kasperlebluebrother: something a bit more powerful than the "iPod Notes" will surely be possible ;)
20:27:11bluebrotheri.e. it should be capable of displaying the manual properly :)
20:27:19Kasperleit's unbelievable
20:27:41Kasperleapple included almost no useful features (e.g. bold or italic text), but <rot13></rot13>
20:27:51bluebrotherpreglow: you should be able killing a ghosted nick with the ghost command to nickserv
20:28:59Febsdircache is definitely the culprit. With dircache off, boot time is < 10 seconds. With dircache on, 2 minutes, 3 seconds. Quite frustrating.
20:29:41 Quit lini ("lini has no reason")
20:29:45 Join tHEkIND [0] (
20:30:12preglowbluebrother: oh, didn't know that
20:30:21amiconnFebs: Then you probably have more files on your unit than dircache can take. Hence it scans on every boot, only to find that you have too many files and it can't initialise
20:30:41tHEkINDI'd like to know the mpegplayer progress about rebuffer, i thought it was an H10-only issue, but it seems to be all targets. Is someone working on it ??
20:30:44FebsAh ha.
20:30:54amiconnIf it would work properly, it would know the necessary buffer size after the first full scan, and would do subsequent scans in the background
20:32:48 Join lini [0] (i=pugsley@
20:34:06*amiconn wonders how many files would cause dircache to scan a full 2 minutes
20:34:27amiconnWhen I tried dircache, scan times were around 20..30 seconds iirc
20:34:31preglowless for me
20:34:39preglowlemme see what stats says
20:34:54Febsamiconn: Looks like you're right. Even after scanning, the dircache info says that the cache is not initialized.
20:35:26amiconnI really don't remember exactly, I don't use dircache during normal operation
20:35:31preglow5063 files, dircache uses about 6 seconds in scan
20:35:56preglowsays "initialized" here
20:35:58preglowand always has, afaik
20:36:23preglowscanning took "8s", it says 5465 entries
20:36:31amiconnJust 6 seconds here on H180...
20:36:47amiconn5578 entries
20:37:06FebsI'm running the stats plugin. It's at 16,000 files and still counting.
20:37:12FebsWhich doesn't seem right.
20:37:41amiconnI'd guess you have >100,000 files if dircache takes that long
20:37:53 Quit Juice^ (Read error: 110 (Connection timed out))
20:38:59amiconnThis MK8007GAH is really damn fast
20:44:13amiconnFebs: If you think stats is wrong, your file system structure might be corrupt. Circular directory link or somesuch
20:44:30amiconnchkdsk should help...
20:45:57FebsAccording to Windows, I have just over 10,300 files, so there is definitely something wrong. I'll try a chkdsk and see what happens.
20:50:22 Join nls [0] (
20:50:29 Part nls
20:51:22Nico_Pamiconn: how can i declare a function that accepts an unlimited number of arguments, in the likes of printf ?
20:51:32 Join Lear [0] (i=chatzill@rockbox/developer/lear)
20:51:41 Join Naked [0] (i=naked@
20:52:22 Join nls [0] (
20:53:32Nico_Phmm i think i found it...
20:53:59Febsamiconn: that fixed it. Thanks. Scan took 17 seconds and now the dircache is initialized.
20:54:11 Join toer [0] (
20:54:22amiconnDid chkdsk tell you what needed fixing?
20:55:10FebsYes. There were bad links in two directories.
20:57:38*amiconn hopes this was not a fat driver bug...
20:57:51amiconnWhich directories? Anything rockbox might have written to?
20:58:55amiconnHmm, not good....
20:59:36FebsActually, though, I don't appear to have a \
20:59:48LearIsn't there a FS bug about an FS bug? :p
21:00:03LearFS = FlySpray and FileSystem, respectively.
21:00:21tHEkINDWould it be hard to code a patch for a new tag on wps -> Activate a tag during X seconds after changing the volume ??????
21:00:34Febs... a \recordings directory (unless it's hidden).
21:01:12FebsI wonder if the issue was caused by me doing a test recording on the ipod.
21:01:33amiconnDid it work?
21:01:55amiconnI think chkdsk deleted /recordings if the bug was in the dir itself
21:02:04FebsCould be.
21:02:20FebsI had tried recording once back in December using a line-in dock that Soap made.
21:02:49FebsIt worked in that I could clearly identify the recorded material, though it was at a level not much above the noise floor.
21:02:58 Nick mcphail_ is now known as mcphail (
21:06:32Nico_PB4gd3r ? LinusN ?
21:06:40 Join Bagder [0] (
21:06:58Nico_PBagder ?
21:08:23Bagderis that a question really?
21:08:51Nico_PBagder: the question is coming :)
21:09:03preglowBagder: click the "iaudio m5" gentlemen post on the rockbox front page and watch the left menu
21:09:05Nico_PBagder: i wanted to know if i could write a function using a variable list of arguments,
21:09:22***Saving seen data "./dancer.seen"
21:10:12BagderNico_P: you mean like sprintf?
21:10:20Nico_PBagder: yes
21:10:35Bagderyes you can ;-)
21:11:10Nico_Pi was asking because it seems to be recent C so i thought maybe it wasn't allowed :)
21:11:25Bagderit isn't recent at all
21:11:41Bagderbut if you speak of macros, it is
21:12:03Nico_Pwhat do you mean ?
21:12:25Bagderfunctions with a variable amount of arguments have been allowed in C since the very beginning
21:12:41Bagderbut macros with a variable amount of arguments have not
21:12:51Bagderthey were added to C in the C99 standard
21:13:04Bagderand have been supported in gcc before that by an extension
21:13:18Nico_Pi was thinking of a function
21:13:32LearAnd such macros are used in Rockbox nowadays.
21:13:56Nico_Pin fact a generic function to get an argument list from a WPS tag
21:14:13Nico_Pbut maybe sscanf is perfect for the job
21:16:02 Quit rp- ("AMS i'm coming, tomorrow :)")
21:16:03LearNot that sscanf is currently included in the core, afaik...
21:16:20 Join RogerBacon [0] (
21:16:46RogerBaconcan someone tell me how to boot the original iriver h10 firmware with the new bootloader ?
21:17:41dan_aRogerBacon: Hold "Left"
21:17:42ArathisRogerBacon: hold down <- while the bootloader apears and until he says something like "loading original firmware"
21:18:05Nico_PLear: it is
21:18:36Nico_Pso i think i should use it instead of reinventing the wheel
21:18:41RogerBaconmaybe im blind, but its not explained in the iriverh10Installation beta wiki page ..
21:18:44Nico_Pespecially a wheel that complicated
21:19:15Learnico_p: Only for some IFP "stub"...
21:19:32LearIt wasn't used by the core, so the doom plugin explicitly links it.
21:19:44Nico_PLear: what's firmware/common/sscanf.c ?
21:19:54LearI went through this when updating bookmark.c.
21:19:57dan_aDoes anyone have any feelings over whether we should make turning "hold" on the way to boot into the OF in the Sansa and H10?
21:20:07LearBut it isn't buildt for the core, see firmware/SOURCES.
21:20:49Arathisdan_a: turning hold on on h10 while booting resets the config
21:21:52dan_aArathis: On the iPods, turning hold on before the Rockbox bootloader (but after turning on the player) boots into the OF, turning hold on during or after the RB bootloader clears settings
21:22:02Nico_PLear: what did you do in the end ?
21:23:15LearSome custom code that did only what was needed. Mainly because I was after saving code at the time. Dragging in a couple of kb (well, lots, compared to what I'd gain) of code wasn't really an option then. :)
21:23:17Arathisdan_a: that may be
21:23:27 Part tHEkIND
21:23:40dan_aArathis: I'm just thinking of making it consistent
21:23:54GodEaterdan_a: sounds like a good idea to me
21:24:01GodEaterI'd be in favour of it (not that I own either player)!
21:24:26 Quit printfXh4 (Read error: 145 (Connection timed out))
21:24:34dan_aGodEater: I thought you had a 5.5G iPod? Or did that get changed to the Gigabeat?
21:24:59Arathisdan_a: okay, when the iriver BL is loading and turning hold on and turning it off before the RB BL loads it doesn't affect RB. but the iriver BL is there for just 1 or 2 seconds
21:25:11 Part K4rP4D ("Leaving")
21:26:41preglowdan_a: any more luck on thumb stuff, btw?
21:27:38LearBtw, regarding bookmarks, I (re)read the bug about the separator used... Any opinions as to what ";" should be replaced with? Tab sounds nice, I think...
21:27:57dan_apreglow: I've not been trying - it's just going to be trial and error to see what works and what doesn't to start with.
21:28:49dan_aI'll see if I can knock 20k off the core tonight as a bit of a challenge!
21:30:53GodEaterdan_a: I meant I don't own an H10 or sansa
21:30:57RogerBacondan_a : thanks, wiki updated
21:31:12GodEaterdan_a: I do own the iPod though - not running RB still though
21:31:29dan_aGodEater: Ah... I don't know where my brain went when reading that!
21:32:46GodEaterdan_a: are you going to conference in to the AMS visit tomorrow ?
21:33:10RogerBaconanother question, the iriver database is constructed with the iriver plus software ?
21:33:31 Quit _Veseliq_ (Remote closed the connection)
21:33:37ArathisRogerBacon: that or with easyh10
21:34:34dan_aGodEater: If there's a call happening, and if I'm not stupidly busy at work, hopefully.
21:35:08RogerBaconwhy using a proprietary database ?
21:35:33eggyhrm, any sugestions on what to do if the ipnod nano (first gen) just goes "dead", it will not respond to anything I do..
21:35:37GodEaterdan_a: I just read the Bagder wanted them to conf him in
21:36:01GodEatereggy: first, charge it for a good couple of hours
21:36:33GodEatereggy: then hard reset it by turning the hold switch on then off, and then holding Menu+Select for about 10 seconds or so
21:37:49eggythere we go
21:39:15pixelmaRogerBacon: rockbox uses its own database if that's what you mean
21:39:58eggyGodEater: whats the combo to get into maint mode/
21:40:36RogerBaconokay, the iriver is in MTP and its not detected by windows (have WMP11)
21:40:47RogerBaconhow i put it in UMS
21:41:07GodEatereggy: after the reset, hold Menu+Play
21:41:13GodEaterassuming you mean disk mode =/
21:41:49ArathisRogerBacon: plug in to usb, boot player and hold down "O" unteil he says "recovery mode"
21:42:04eggyi just forget all those special options, been a bit sense i've played with ipodlinux (though this seems a hell of a lot nicer)
21:43:00 Quit GodEater_ ("CGI:IRC (Ping timeout)")
21:43:33RogerBaconokay, but its possible to use a UMS bootloader, because the h10 is MTP only (unable to put it in UMS, the european firmware dosn't work)
21:44:17ArathisRogerBacon: mine is also mtp and using this ums trick I can connect it to the pc
21:45:13ArathisRogerBacon: the player needs to be reset before doing the ums trick. for 5/6gb plug out battery and put it in again. for 20gb hit the reset button on top of the player
21:46:31RogerBaconArathis : okay thanks
21:47:47RogerBacondid you try using the iriver plus or easy h10 to build the database (dont know if the device is detected with this trick by iriver plus ?)
21:48:01dan_aWTF???? preglow: I don't think I'll be doing any thumb code work tonight - GCC is compiling gode which talks to non-existant registers
21:48:35ArathisRogerBacon: it isn't by the iriver plus software I think. I used easyh10 when using the OF back
21:49:29RogerBaconokay, if i put file directly in music, the database will be empty ?
21:49:52pixelmaRogerBacon: are you talking about the database you want to use in rockbox?
21:51:09ArathisRogerBacon: rockbox has it's own database. The one from the original firmware doesn't affed it
21:52:23RogerBaconokay thanks guy's
21:52:47 Quit preglow (Read error: 104 (Connection reset by peer))
21:55:21 Quit blueworm ("Leaving")
21:56:28 Join GreyFoux [0] (
21:57:19 Join preglow [0] (
21:58:51 Join blueworm [0] (
21:58:59RogerBaconso, any news form the zune-linux guy's (they said that they will contact us) .... :)
21:59:29*RogerBacon is really sarcastic ...
21:59:58*Bagder giggles
22:00:19preglowthey've been really silent...
22:00:43preglowBagder: did you get my comment about the left menu?
22:00:59BagderI'll poke on that later
22:01:23amiconnBagder: Could you perhaps also fix
22:02:50Bagderyou mean "fix" as in remove some stuff from archos zips?
22:02:54 Quit GreyFoux ("Le vrai danger, c'est quand les hommes penseront comme les ordinateurs")
22:03:22 Quit muesli__ (Read error: 104 (Connection reset by peer))
22:03:29amiconn(1) Don't include eq settings for hwcodec. (2) Don't include tha 'backdrops' dir for monochrome targets
22:04:40amiconn(2) is not only archos, but also e.g. iFP (though is not yet actually used for iFP)
22:05:29 Join muesli__ [0] (n=muesli_t@
22:08:41 Join robin0800 [0] (
22:11:07 Join webguest01 [0] (i=436769f2@gateway/web/cgi-irc/
22:12:46 Join groovingandi [0] (
22:13:39 Join ender` [0] (
22:13:41 Quit webguest01 (Client Quit)
22:16:03 Nick hcs is now known as hcs|mountaineeri (n=agashlin@rockbox/contributor/hcs)
22:16:45 Nick hcs|mountaineeri is now known as hcs|pillaging (n=agashlin@nat01-silvers-ext.Rutgers.EDU)
22:22:59 Join Alonea [0] (
22:23:51groovingandiI just tried a recent daily build with my Archos Recorder 8MB (haven't updated for a long time)... Does anybody know, why the key combos of the quick screens were disabled?
22:24:20groovingandiOn+play in the file browser doesn't work anymore either
22:25:43 Quit Lear ("Chatzilla 0.9.77 [Firefox]")
22:27:26amiconnOn+Play is gone for a long long while. It's long Play instead
22:28:21groovingandiI know that "long play" works as well and I think I'll get used to it... what about the quick screens?
22:30:10amiconnHmm, not sure. They probably get lost in the button actions move and so far nobody complained they're missing
22:30:19*amiconn points to JdGordon
22:30:23groovingandiBtw. I think I like the new menu structure
22:30:26amiconnNot around atm...
22:30:51groovingandiHehe, probably I'm the only one who is still using an old Archos box
22:31:01 Join mindmedic [0] (
22:31:06amiconnNah, surely not
22:31:18amiconnI'm still using my archoses as well as the newer boxes
22:31:26amiconnBut I never used the quickscreens much
22:32:15pixelmanls: I'm trying to get familiar with the langV2 patch... I've found a few (seen 9 so far) sound settings that are currently defined for "hwcodec" but the Player which has a different MAS does not have. Do you think it's worth it do invent a new category for these?
22:32:54*amiconn would think so
22:33:01 Join BobJonkman [0] (
22:33:02nlspixelma: I guess it's a good idea...
22:33:25 Quit obo ("bye")
22:33:27nlsI wasn't sure about what features which MAS has.
22:33:47 Quit kubiix (Read error: 104 (Connection reset by peer))
22:33:49amiconngroovingandi: The archoses still have a number of advantages over all these newer boxes
22:34:19 Join obo [0] (
22:34:27nlspixelma: I have a slightly updated patch that has been synced with svn here, if you want, no big changes over the last one in the tracker.
22:34:30amiconnOf course they also have disadvantages, it all depends on usage patterns
22:35:34pixelmaalright... thanks for the opinions. Well it's easy to write that in the .lang now to find out what else needs changing and how to write that in the features.txt
22:36:09pixelmanls: I'd be interested
22:36:09 Quit Arathis ("Bye, bye")
22:36:33nlspixelma: ok, I'll just check that it's still in sync and post it :-)
22:36:35groovingandi@amiconn: I have 2 Recorders, an Ondio SP and a H340 but use my first recorder most of the time
22:36:52pixelmanls: thanks :)
22:37:12amiconnI also use the my Recorder 80 quite often, as well as my OndioFM
22:37:35groovingandisomehow I got never used to the keys of the H340, although I really like the size and the display
22:37:41 Quit webtaz ("( :: NoNameScript 4.02 :: )")
22:38:05groovingandiand of course 40G are simply not enough...
22:38:23amiconnI'll try to switch to my H180 as my main box. Now I have the advantage to put my cd collection as flac :)
22:39:27 Quit RogerBacon ("Quitte")
22:40:11nlspixelma: ok, synced patch posted
22:41:08 Quit mindmedic ("Leaving.")
22:42:19 Join FOAD_ [0] (
22:43:45dan_aI'm looking at improving the Sansa flash driver. Should it yield every time it is doing a busy wait?
22:44:24pixelmanls: stupid question - what should be done with the duplicated lang strings (like "all" for files and repeat mode)? Will that be cleaned up when everything else is done?
22:44:50 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:45:00 Quit amiconn (Nick collision from services.)
22:45:02 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:47:36groovingandiamiconn: I think the On+Up/Down combos for the pitch screen were also lost
22:48:01*amiconn doesn't even know what these did
22:48:45 Join tucoz [0] (
22:48:56groovingandiIt was possible to change pitch up/down in the WPS
22:49:33amiconnAh, in wps...
22:50:47groovingandiOn+Play in WPS for muting doesn't work either...
22:51:01groovingandiI never used it... just testing different combos now
22:51:12pixelmathere were changes for the pitch screen - about 2/3 down
22:51:18amiconnHehe, funny how much I didn't even know... although I'm using rockbox for almost 5 years now, am developing for it for more than 4 years, and have 8 targets...
22:51:29groovingandisorry, F1+play for muting
22:51:40amiconnMute was removed
22:52:06groovingandiOk, didn't know it existed until 5 mins ago
22:52:18groovingandiI found it in the manual
22:52:27amiconnDoesn't really make sense, pause is way more appropriate
22:52:40amiconnThe manual still needs a lot of work...
22:53:23pixelmathere's a "Pitch" paragraph - don't know how accurate that is though... but I think this one is newer than the WPS key table
22:54:07groovingandipixelma: yes, but I can't access the pitch screen from WPS
22:54:47amiconnAh, now I remember what you mean with On+Up/Down
22:54:53amiconnThat should work, in fcat
22:55:27groovingandiI just saw it works from the WPS context menu, there's a pitch entry
22:55:36amiconnOn+Up or On+Down should bring it up
22:56:04groovingandidoesn't work for me
22:57:21 Join kubiix [0] (
22:58:09 Quit FOAD (Read error: 110 (Connection timed out))
22:59:37 Join Arathis [0] (
23:03:45groovingandiI just downloaded the sources...
23:03:58groovingandiin keymap-recorder.c it's defined:
23:05:38 Join dxm [0] (
23:05:50amiconnAnd another one for On+Down
23:05:54groovingandidon't know why it doesn't work
23:06:18dxmwould someone explain to me how to install the dictionary plugin i have all the files
23:06:30groovingandiACTION_LISTTREE_PGUP works
23:07:40 Join BHSPitLappy [0] (
23:07:40 Part perl|work
23:09:26***Saving seen data "./dancer.seen"
23:11:50dxmhas anyone succesfully installed the dictionary plugin
23:12:30groovingandidxm: sorry, never tried it
23:13:01 Join petur [0] (n=petur@rockbox/developer/petur)
23:13:48dxmthe directions say to put some files in a folder and "execute"−−how do you do that?
23:14:02amiconnThe dictionary plugin is part of the official distribution
23:14:24amiconnYou just need to add the dictionary you want
23:14:39amiconnFor this perl script you need a perl interpreter
23:15:17dxmwhat would be an example of a perl interpreter
23:16:29dxmare these perl interpreters user friendly? obviously i know nothing of the sort
23:16:51dxmcan perl interpreters "execute" .pl files?
23:16:59bluebrotherdxm: it's an interpreter. It simply runs the script file.
23:17:09 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
23:17:23groovingandiyou are using windows?
23:17:29dxmyes windows
23:17:34bluebrotherno gui, no clicking ...
23:17:47bluebrotherbut perl has modules for creating gui stuff
23:18:25dxmmaybe if i get a perl interpreter right now you could walk me throught it? :)
23:18:36dxmit seems simple enough if you know perl
23:18:41bluebrotherlook for Activestate Perl
23:20:30 Nick hcs|pillaging is now known as hcs (n=agashlin@nat01-silvers-ext.Rutgers.EDU)
23:20:36 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
23:20:47dxm30 megs
23:21:37nlspixelma: I have tried to clean up all duplicated strings I have found but some still needs to be duplicated, such as "All" because it is used in the quick screen which relies on the sysfont when the userfont doesn't fit. If you find other duplicates that can be removed, then go for it :-)
23:22:01groovingandidxm: ActivePerl is enough, you don't need a development environment
23:22:39dxmok now "activeperl"
23:23:16pixelmanls: thanks for the explanation
23:23:36dxm"as package" or msi?
23:23:49 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
23:24:35groovingandiboth should work, I guess msi is easier (MS installer)
23:25:24 Quit Alonea ("Leaving")
23:25:29dxmbtw thank you for helping me
23:26:25 Quit tucoz ("Leaving")
23:26:29safetydanoh bah... now I know why speex stereo decoding doesn't work
23:26:58 Quit OgMaciel ("mv OgMaciel $HOME")
23:27:16 Quit funky ("leaving")
23:29:02dxmstep one is to get a dictionary file, which i did. the next step is to extract and from it
23:29:07 Join Ribs [0] (n=ribs@
23:29:49 Quit robin0800 ("A day without sunshine is like .... night")
23:31:34 Quit Febs (Read error: 110 (Connection timed out))
23:32:37groovingandidxm: you are following the instructions of the wiki page?
23:33:04dxmfor converting the dictionary to rockbox readable format
23:33:18groovingandiso you downloaded WNprolog-2.0.tar.gz
23:33:38groovingandiwere you able to unpack it?
23:33:58dxmwait yes
23:34:10dxmwinrar does it
23:34:24dxmoh so i pull those files out of there...
23:34:51groovingandiyou should know that I've never done this before, so I don't know what I'm talking about
23:35:04 Part hcs
23:35:31dxmyou've got me alot further than ive ever been
23:35:55dxmbut what about this
23:35:58dxmwhere is that?
23:37:32groovingandiit's in the rockbox sources
23:37:42groovingandidid you read until the bottom of the wiki page?
23:37:44dxmi googled it and found it in the rockbox sources. ok so
23:37:54dxmhaha yes
23:38:03groovingandiyou can download the dictionary files, no need to build them yourself
23:38:18dxmhaha no
23:38:26dxmthere are the files right there eh?
23:39:10dxmsilly me thanks for letting me waste your time happy hunting
23:39:19 Quit jhMikeS (Nick collision from services.)
23:39:25 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
23:40:29groovingandino problem, so you don't need perl anymore
23:40:34dxmbut.. do you want to do it anyway just for kicks?
23:40:41dxmi love learning
23:41:04preglowmakes for a fun life
23:41:13groovingandishouldn't be difficult, just install activePerl and download
23:41:25 Join entheh [0] (
23:42:23dxmim sure you must be right but i have activeperl installed and open and i haven't a clue about what im seeing its like another language! haha
23:42:50dxmpackages are pl files?
23:43:09 Join muesli- [0] (n=muesli_t@
23:43:18groovingandijust open a command window a type "perl -v" to check it
23:43:44dxmah so from the command line i can execute pl files now?
23:44:04groovingandiif installation was successful, yes
23:45:02dxmis the rdf2binary tool in the rockbox sources
23:45:10dxmwhat filetype will that be?
23:46:43dxmcant seem to find it with google
23:47:29groovingandiin the tools directory of the rockbox sources, there's rdf2binary.c
23:47:31nlsdxm: I suggest you take a look in the tools dir in the rockbox sources
23:47:37dxmyes thanks
23:48:43dxmim sure you must think this is quite ridiculous but where are the rockbox sources
23:48:52groovingandiyou have to compile it first, don't know if it works for windows
23:49:45dxmis there a program i can use to complile it?
23:51:30dxmok so is compliling done
23:51:44dxm*how is compiling done
23:52:58groovingandiyou need a c compiler, gcc for example or the MS variant
23:53:05 Quit muesli__ (Read error: 110 (Connection timed out))
23:53:37 Quit SirFunk (Read error: 110 (Connection timed out))
23:54:10 Quit Ribs ("eh eh ehhhh!")
23:54:21groovingandiif you don't have one installed, it's probably really to much effort
23:54:35dxmah you're probably right
23:54:56dxmalthough i wonder what could be so time consuming about it
23:55:14groovingandiproblem is that I can't help you with compiling as I have to leave soon
23:55:52dxmok thanks again i cant wait till i have another thing i want to do thats way over my head with rockbox
23:55:53groovingandiand I can't test what is necessary for windows
23:56:13 Quit petur ("Zzzzz")
23:56:21nlsdxm: take a look at the SimpleGuideToCompiling to set up a dev environment for rockbox deleopment, if you only want to compile this secific tool, it is a bit overkill, as any c compiler should do
23:56:54 Quit desowin ("use linux")
23:56:58dxmsetting up a dev environment for rockbox dev...that does sound like alot of work
23:57:20nlsnot if you use the vmware image
23:57:35nlsinstall vmware, download image, and go
23:57:41groovingandior you can try if you find a windows installer for gcc
23:58:04nlsgroovingandi: I think that is known as MingW or somesuch
23:58:53groovingandiyeah, MingW or a Cygwin environment

Previous day | Next day