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

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

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

#rockbox log for 2006-05-09

00:00:22 Join MagusG [0] (
00:00:40amiconn...and perhaps even feed analog input signals to s/pdif out
00:00:57preglowamiconn: echo spdif in to spdif out?
00:01:11preglowwhere would we get a proper clock?
00:01:21amiconnErm, from input?
00:01:33preglowright, sclk1-4
00:01:36 Quit dpro (Connection timed out)
00:01:37preglowi thought those were the iis clocks
00:01:44preglowbut they're not, obviously
00:02:00preglowwell, sure, not a bad idea
00:02:05 Join Rincewind_12 [0] (
00:02:34preglowi'd actually say it's a good one, so agreed
00:02:41preglowi can't test it, though
00:03:12amiconnI can't test it as I would need a second toslink cable + a second toslink->mini adapter
00:03:32preglowi'd need another spdif input capable device
00:03:34preglowwhich i haven't got
00:03:43amiconnThen I could feed dvd player -> h140, and h140 -> opt/coax converter -> archos recorder
00:03:45preglowi have the extra cable
00:04:31preglowbut no
00:04:35amiconnI'll try your s/pdif recording next
00:04:45preglowgo on
00:04:51preglowi'll try to update the patch tomorrow
00:04:55preglowi don't have my player here now
00:05:03preglowso can't test
00:05:11 Join Genre9mp3 [0] (n=Yngwiejo@
00:05:23amiconnThe analyzer shows symbol error with no signal attached, an when pulling the plug half-way I also get parity error (of course)
00:06:09preglowif you discover why dma suddenly stops when unplugging several times, i'll be very happy
00:06:12preglowi have no idea
00:08:52 Join m0nk_ [0] (
00:08:58m0nk_hey guys
00:09:32 Quit Genre9mp3 (Client Quit)
00:09:47markunm0nk_: welcome back
00:09:54m0nk_thank you mark
00:10:11markunm0nk_: we have a #gigabeat btw..
00:10:35 Join Genre9mp3 [0] (n=Yngwiejo@
00:10:41m0nk_awesome i will check it out laterish
00:10:42amiconnpreglow: This is somewhat strange. (1) I get no monitoring although I think I should?
00:10:55m0nk_i wanna see whats going down here for now
00:11:05preglowamiconn: indeed you should
00:11:17amiconn(2) When unplugging, the size counter keeps running. After repeated plug/unplug, recording stops
00:11:32preglowyep, that's what i get too
00:11:36amiconnI think this is normal behaviour, as the recording dma isr checks for dma errors
00:11:37preglowbut the monitoring should work, afaik
00:11:54m0nk_grr i need a hot chik to goto the movies with:|
00:12:02amiconnBut: if I then start recording again, it hangs after 8KB (1 dma block)
00:12:04m0nk_I havent been out with a girl in too long
00:12:15preglowamiconn: exactly
00:12:23preglowamiconn: you need to exit, then reenter the recording screen
00:12:25preglowthen it'll work again
00:12:25 Quit egtrippen ("CGI:IRC (EOF)")
00:12:33preglowso something in the init code fixes the situation
00:13:08amiconnI can't, rockbox hangs
00:13:21preglowyes, before you try to record again
00:13:27preglowat that point, it's too late
00:14:07amiconnI'm still puzzled by the fact that the transfer keeps going with no signal
00:14:14preglowyes, it's weird
00:15:46 Quit BHSPitMonkey__ (Read error: 104 (Connection reset by peer))
00:16:49 Join Daishi [0] (
00:17:15amiconnHmm, now monitoring works???
00:17:35preglowdid you just rolo the new firmware?
00:17:48preglowthe first time, that is
00:17:48 Part m0nk_ ("Leaving")
00:19:10preglowthen that explains it, probably
00:19:46preglowhardware regs not being left in the state i expect them to be, i guess
00:20:07amiconnBtw, DMA only drops out on parity error. When pulling s/pdif fast, it keeps going, but when pulling slow, causing parity errors, it drops
00:20:29preglowyou sure? i thought i noticed parity errors when the dma didn't drop as well
00:20:39preglowi've hardwired the status display to show me the interrupt status flag
00:20:46amiconnIf so, that needs investigation. There should be no difference between rolo and boot
00:21:28preglownot much work has been done with spdif init
00:21:49preglowso some regs might be left in some weird state
00:22:15 Quit MagusG (Connection timed out)
00:22:28 Quit midkay (Read error: 104 (Connection reset by peer))
00:22:44 Quit petur ()
00:22:45 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:27:14Genre9mp3When you have status bar set to off...the expected behavior for the menu/browser text is to go on top of the screen right?
00:27:21 Join MagusG [0] (
00:29:20 Join Thus0 [0] (n=Thus0@
00:29:41Thus0I've just installed rockbox on my iriver :)
00:30:03XavierGrGenre9mp3: yes
00:30:45 Join Paul_The_Ner1 [0] (
00:31:08Genre9mp3Well, at least this happens in the main LCD.....but in my H300LCD Remote this doesn't happen. When the status bar is off, the text remain at its initial position, and in the area where the status bar is you can see part of the WPS
00:31:16Genre9mp3this is a bug, right?
00:31:34 Part Paul_The_Ner1
00:32:17preglowThus0: then your life just took a turn for the better
00:32:53XavierGrGenre9mp3: Fill a bug report on the bug tracker. I will try to look at this tomorrow
00:33:21Genre9mp3I suppose this happens for both the H100 and H300 Remotes
00:33:37 Join TCK [0] (
00:33:57XavierGrmore probably
00:34:43Genre9mp3Also, when I start the player, the staus bar area in the remote is blank....
00:36:18XavierGrGenre: In my H100 when I turn of the status bar the text moves up
00:38:12 Quit gursikh (Read error: 110 (Connection timed out))
00:38:22 Join sockerteze [0] (
00:41:30 Join Dinomight [0] (
00:43:39Dinomightdoes anyone have a link to some eq presets (i'm not an audiophile and am not really certian how to make good ones for standard music types (ie rock, classical, pop, metal)
00:43:48preglowwe haven't made any
00:43:55preglowperhaps the mistic creq has
00:44:15 Join RoC_MM [0] (
00:44:48Genre9mp3and I think that you should try making on your own....because these things are subjective
00:44:55Galoisalso it would depend on your headphones and such anyway
00:45:37RoC_MMNote to all who didn't hear last night: iPod 4G Grey 5/7 build is not functional and does not get beyond boot logo. 5/5 build works.
00:46:13preglowof course your should make them on your own, but it's still fun to have presets
00:46:46Genre9mp3I know that someone made 5 presetes in the MR
00:47:06XavierGrGenre on my H100 your WPS fits perfectly on both screens (remote, main) with and without the status bar.
00:47:24Rincewind_12pilot's experimental build for h100 has some eq presets too
00:47:35amiconnGah, this is weirdo
00:47:57amiconnI think I know where the 8KB come from
00:48:02 Join gtkspert [0] (
00:48:18Genre9mp3XavierGr: The problem is not in the WPS screen....just set the status bar off and the go to the WPS and then back to menu...
00:48:36preglowamiconn: hwere?
00:48:53amiconnThe pcm recording is simplistic. A bit too simplistic. It _assumes_ the transfer succeeded when receiving the DMA interrupt
00:48:57preglowfirst person to make fun of my lousy typing gets killed
00:49:03XavierGrGenre9: I just did
00:49:07Genre9mp3XavierGr: Do you get the text in the first line there? I don't
00:49:11XavierGrand nothing seems wrong
00:49:18XavierGrI will try on H300 too
00:49:20amiconn...but the interrupt doesn't necessarily mean it did. It can also mean 'error'
00:49:27 Join scott666 [0] (
00:49:37preglowthat's worth checking for sure
00:49:43preglowbut why would it fail?
00:49:52Genre9mp3XavierGr: maybe this is only for the H300 LCD Remotes??
00:50:22amiconnIt looks like the recording dma is supposed to run all the time while in recording screen, even if no recording is going on
00:50:49preglowit is
00:50:51preglowfor the peak meter
00:51:09preglowwhich is always on
00:51:38amiconnA stop-from-error posts PCMREC_STOP, and pcmrec_stop() in turn calls pcmrec_dma_start();
00:52:24amiconnI think it's already this restart which fails, but due to the simplistic logic the recording code thinks one full block was transferred
00:52:35 Quit Rincewind_12 (Read error: 104 (Connection reset by peer))
00:52:37preglowmakes sense
00:52:59 Quit sockerteze (Read error: 131 (Connection reset by peer))
00:53:08amiconnI added INTERRUPTCLEAR = 0x03c00000; to pcmrec_dma_start() for a test, but to no avail
00:54:15 Quit ender` (" Be wary of strong drink. It can make you shoot at tax collectors and miss.")
00:54:55XavierGrGenre: Are you sure? I just checked with my H300 and all seems to be fine
00:56:51 Join Paul_The_Nerd [0] (
00:58:03 Quit Dinomight ("CGI:IRC")
01:00:40amiconnpreglow: Did you test with logf?
01:00:40PaulJamisn't the cpu supposed to stay boosted all the time when the codec-buffer gets refilled?
01:00:58Paul_The_NerdIt's supposed to boost while filling, yeah.
01:01:18preglowamiconn: i have, but it depends for what you mean
01:01:19Paul_The_NerdAt least it *used* to be supposed to.
01:02:24PaulJamon my device it changes between 124 and 45Mhz. but i will check again with a cvs build.
01:03:45Paul_The_NerdWell it's also possible the buffering behaviour's changed recently. I wouldn't really be even close to an authority. ;-)
01:04:39PaulJami just wanted to mention it, in case this isn' right
01:05:21amiconnpreglow: As I suspected: dma error fires twice in a row, effectively disabling dma
01:05:52preglowamiconn: and again, only on parity error?
01:06:27amiconnNot sure, it happens when pulling/plugging
01:06:42amiconnThe weird thing is that it says DMA error 0x41
01:06:51preglowset the sample rate: display in recording.c to display the interrupt status register
01:06:54preglowit helps
01:06:59preglowat least you can see just what triggers it
01:10:03PaulJamwith a cvs build it also unboosts while filling the buffer.
01:11:20 Quit MagusG (Connection timed out)
01:11:32amiconnHow the heck can the interrupt fire twice in a row with dma error??
01:12:40 Join BHSPitLappy2 [0] (
01:13:12 Quit BHSPitLappy (Nick collision from services.)
01:13:22 Nick BHSPitLappy2 is now known as BHSPitLappy (
01:13:26 Join Farpenoodle [0] (
01:13:39amiconnHmm I think the DMA1 isr has a timing problem....
01:13:57amiconnIt clears the error flags before determining whether to stop dma dma continues running
01:14:19 Join MagusG [0] (
01:19:54 Quit scottder (Read error: 110 (Connection timed out))
01:20:26 Join ashridah [0] (
01:21:24 Join qwm [0] (
01:24:20 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
01:25:24RoC_MMWhat is tentative date for 3.0?
01:25:40preglow15. or 29. i thiunk
01:30:19amiconnpreglow: What's irritating me is that the doubled dma_stop doesn't trigger a dma_start, even though it should
01:30:21 Quit Thus0 (Client Quit)
01:30:31amiconnNo "dma1 started" in the log
01:30:42preglowwhy is it stopped in the first place?
01:31:11amiconnbecause of the DMA error...
01:31:18preglowwhy is there one?
01:31:35preglowdma should just read from PDIR2, no? why does it care it's filled with invalid data?
01:31:55 Part Paul_The_Nerd
01:32:14amiconnNot 100% sure. My guess is that PDIR2 does report something wrong when there's an error
01:32:22 Quit Gargamale ("poop")
01:32:26 Quit mikearthur (Read error: 104 (Connection reset by peer))
01:33:18amiconnAhahaha, it can't!
01:33:55amiconnpcmrec_stop() checks whether recording is already stopped: if (!is_recording) ...
01:34:03***Saving seen data "./dancer.seen"
01:34:06preglowoh, right, that
01:34:10amiconn...but the isr sets is_recording to false in case of error
01:34:32 Quit qwx (Read error: 110 (Connection timed out))
01:35:29amiconnDMA overrun will cause similar behaviour
01:39:03 Join mirak [0] (
01:40:15 Join BHSPitLappy2 [0] (
01:43:42 Join jbauman [0] (
01:45:08 Quit BHSPitLappy (Nick collision from services.)
01:45:24amiconnBoth cases (DMA error and DMA overrun) will cause the DMA not to be restarted, leading to a hang next time recording should strat
01:45:29 Quit _Lucretia_ (Read error: 110 (Connection timed out))
01:45:51amiconnThis is clearly visible by the non-working peakmeter
01:45:58preglowbut yes
01:46:09preglowif we notice an error this way, we should just stop recording
01:46:26amiconnwhich way?
01:46:27preglowand use that as the error condition instead of spdif parity/symbol error
01:46:32 Quit dpassen1 ()
01:46:36preglowdma being stopped
01:46:51 Join _Lucretia_ [0] (
01:46:51amiconnI'd rather restart dma
01:47:03preglowcan you be sure the error condition isn't still valid?
01:47:09preglowit gets stopped for a reason
01:47:32amiconnIn case of overrun: yes, there is a real reason: data not flused fast enough
01:47:46amiconnIn case of DMA error: No real reason, imho
01:47:59preglowit's real enough, apparently
01:48:05pregloweven though i don't know what triggers it
01:48:18amiconnOn archos, I can fiddle with the signal no matter what, it sometimes records garbage, but it neither stops nor hangs
01:48:34preglowwell, sounds good enough
01:48:43preglowif you're yanking the cable in and out, you get what you want
01:48:45amiconnIt just halts transfer with no signal, but that's all
01:48:50 Quit XavierGr ()
01:49:18amiconnIt's handled by the mas - it simply doesn't encode without a signal in s/pdif mode
01:49:28 Quit Moos ("Glory to Rockbox !!!")
01:49:35preglowbtw, i just remembered i actually do have another spdif in
01:49:42preglowmy old minidisc player
01:49:48preglowi just need another tos/jack adapter
01:49:57preglowi'll see if i can find a shop that carries them around here
01:50:26amiconnI know a shop here, maybe I'll visit it a second time tomorrow (for a cable + adapter)
01:50:28preglowno, two tos/jack adapters...
01:51:33amiconnThe archos recording engine doesn't even stop on buffer overrun. It's debatable whether this is good or not
01:51:44preglowwhy not?
01:51:50preglowit should signal an error, but also keep recodring
01:51:53preglowrecording, even
01:52:32amiconnThe recording transfer happens in an isr. It doesn't care whether the thread flushes in time or not
01:52:49 Quit webmind (Read error: 104 (Connection reset by peer))
01:53:26 Join akaidiota [0] (
01:53:27amiconnThat lead to strange effects, like the second buffer round overwriting the beginning of the first, in case the first flush was delayed for too long
01:53:38preglowbut yeah, i can't think of any reason why we wouldn't want to echo the spdif in signal
01:53:41preglowso we should do that
01:53:45 Quit Benacool ()
01:54:37 Join webmind [0] (
01:54:46amiconnThere was a combination of conditions where that would happen: (1) the peakmeter drawing cpu power by running in high-perf mode and (2) the fat driver needing to search for the next free block, in case the nextfree hint was erased
01:55:05 Quit tucoz ("Leaving")
01:55:13amiconnI solved it by avoiding (1) - switching peakmeter to low-perf while flushing to disk
01:55:49amiconnStopping (or waiting) for the flush to finish would require a framewalker, which we still don't have
01:58:44 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
01:59:13preglowbut yeah, what about recording with no spdif in, then?
01:59:33preglowi can see no reason it shouldn't be allowed if we do proper dma restarting all the time
01:59:52preglowcurrently recording hangs sometimes if you record with no spdif source connected
02:00:04preglowand the iriver fw doesn't let you record if you do that
02:00:05preglowshould we allow it?
02:00:14amiconnPerfect silence...
02:01:12 Quit tvelocity ("Ex-Chat")
02:02:07amiconnHmm, for some reason the DMA doesn't like just resetting the error flag
02:03:12 Quit Seed (Nick collision from services.)
02:03:18 Join Seed [0] (
02:04:57 Join whatboutbob_ [0] (
02:05:11RoC_MMis recording through line-out/in a secondary priority or what is the gist of that?
02:05:12 Join JdGordon [0] (
02:09:11 Join mikearthur [0] (
02:10:54 Quit akaidiot (Read error: 110 (Connection timed out))
02:11:03preglowrecording through line-out ???
02:11:50RoC_MMthe eighth inch jack
02:11:53RoC_MMthat headphones go in
02:11:58RoC_MMrecording in through that
02:11:58preglowwhat target?
02:12:04RoC_MMiPod in general.
02:12:04 Quit mikearthur (Client Quit)
02:12:14 Quit MagusG (Connection timed out)
02:12:21RoC_MMit's "line-out" if it's a headphone plugged in, and "line-in" if a mic is plugged in
02:12:30RoC_MMor so I would assume the language goes
02:13:03preglowrecording on ipod isn't much priority for me, at least
02:13:08preglowdon't know about the other ipod coders
02:13:15RoC_MMthat's what I'm looking for.
02:13:18preglowhell, it's no priority at all
02:13:23preglowdon't even know if it's possible on my nano
02:13:31RoC_MMsorta as a "neat feature"
02:13:41RoC_MMI have a use for it if it comes too
02:14:01RoC_MMwhat would be the problem on the nano?
02:14:19 Join MagusG [0] (
02:14:40whatboutbob_preglow/amiconn: sounds like you've found why spdif recording hangs?
02:16:05preglowsounds more like amiconn has
02:16:10preglowi don't even have my h120 here
02:17:05amiconnThere's a strange thing: The DMA error reports as 0x41, which means 'configuration error', i.e. the transfer size and/or addresses don't match the data size
02:17:29amiconn(not integer multiple of bytes, or not aligned)
02:17:50amiconnBut I displayed, SAR1, DAR1 and BCR1.
02:18:23amiconnThey look fine, but BCR1 is zero (apart from the upper 8 bit which are reserved)
02:19:11amiconnSo it seems the DMA error is delayed until the actual end-of-transfer (unless the controller fiddles with these values on error, which I don't think it does)
02:21:02preglowsounds unlikely
02:22:08amiconnI just hard-crashed the unit...
02:22:39amiconnIt switched off and started to reboot into iriver - which hung at the start screen, ata led on
02:23:27preglownot bad...
02:23:45preglowhow'd you manage that?
02:24:05amiconnI just tried to start the next chunk in case of DMA error...
02:24:18 Quit jaebird ("Ex-Chat")
02:25:16 Quit ashridah ("Leaving")
02:26:30 Quit Acksaw ("I'm off, see ya later!")
02:27:03 Quit Farpenoodle (Read error: 110 (Connection timed out))
02:27:29 Quit PaulJam (".")
02:27:35amiconnHmm, something overflows if the interrupt fires too often (but I had it working for quite a while with heavy errors)
02:27:48amiconnThen got: Panic: No Font
02:28:04 Quit Genre9mp3 ("Leaving")
02:28:59amiconnprobably due to logf()ing in the isr
02:29:36amiconnyeah, I'm giving it a hard time...
02:29:56preglowit does number among the things one should not do in an isr, yes
02:30:02preglowbut hell, i need to sleep
02:30:13preglowi'll be reading the logs if you discover anything more
02:30:16preglowgood night
02:30:18amiconnFinal test, w/o logf
02:31:40 Quit Seed (Nick collision from services.)
02:31:47 Join Seed [0] (
02:31:59amiconnIt works! It's teh hack, but it works :)
02:36:47whatboutbob_nice one!
02:37:46 Part synth7
02:38:27amiconnIt currently forces the DMA to run again. We'll need proper handling, but now we know what happens, and how to recover
02:39:03amiconnOf course such a recording won't be glitch-free,
02:39:37amiconnbut in fact I think we should continue recording when there is an error. Better a slight glitch than a stopped recording...
02:42:20amiconnOf course it would be better to trigger on the symbol error/ parity error interrupts, and stop DMA until the errors go away
02:48:21whatboutbob_amiconn: i agree. better an errored something than nothing.
02:48:42 Join qwx [0] (
02:50:58 Nick BHSPitLappy2 is now known as BHSPitLappy (
02:51:15 Join XavierGr [0] (
02:51:47 Join perl|wtf [0] (
02:58:14 Quit qwm (Read error: 110 (Connection timed out))
03:02:02sharpei shake.
03:02:24perl|wtffreezing period?
03:04:52 Join sockerteze [0] (
03:06:19 Quit pixelma (" HydraIRC -> <- Get hot chicks here!")
03:07:29sharpeyou know, it may be due to the amount of caffeine i usually consume...
03:09:39whatboutbob_amiconn: let me know if there's a new patch you'd like me to test.
03:09:54sharpeokay, i'll get to work on the emulator this weekend.
03:11:45 Quit sockerteze (Read error: 131 (Connection reset by peer))
03:13:51 Quit jbauman (Read error: 104 (Connection reset by peer))
03:15:58 Join sockerteze [0] (
03:22:38 Join dsh-1 [0] (
03:24:00RhapsodyCan I get a different OS to boot by default?
03:24:21Galoisyou have more than two?
03:24:58 Quit sockerteze ("Chatzilla 0.9.73 [Firefox 1.0.7/20050915]")
03:25:13RhapsodyI would have liked to have three but I had to delete one.
03:25:25RhapsodyCan I set the iPod firmware as the default OS to boot?
03:26:06GaloisI wrote a patch for this, which is surprisingly popular until people realize that they'll have to compile it themselves because I won't do it for them
03:26:29RhapsodyNo problem for me. ;-)
03:26:39RhapsodyWill it interfere with a CODEC patch I use?
03:26:41 Quit Daishi (Nick collision from services.)
03:26:44 Nick dsh-1 is now known as Daishi (
03:26:56Galoisno, it's one line and it's in the bootloader file which has nothing to do with codecs
03:27:21RhapsodyOkay, I'll run it when tonight's source is released.
03:29:20RhapsodyWhat button do I hold in yours to boot Rockbox?
03:29:51RhapsodyAlright, I just read the C.
03:30:34RhapsodyI would have preferred not to have to compile Rockbox at all but this is no problem because I already have to recompile for the CODEC I want.
03:30:51Galoisoh and there's no reason to wait for tonight's source, this patch is in the bootloader, and the bootloader is different from the usual
03:31:16Galoisyou have to reinstall the bootloader to use the patch (see, it gets complicated already)
03:31:23Galoisthat means ipodpatcher blah blah
03:31:30Galoisipod_fw etc.
03:31:40RhapsodyOh, that.
03:31:48Galoisthe bootloader hasn't changed in like a month
03:31:53RhapsodyAlright, I'll find the source to that.
03:32:32Galois../tools/configure ; <select ipod model> ; <press B for bootloader build> and so on
03:33:11RhapsodyYes, I am familiar with it.
03:33:17dongsoh hes back.
03:33:53RhapsodyI need to find the correct packages so I can run it on my Mac, right now I had to do it using Cygwin on an old PC I had lying around.
03:34:07***Saving seen data "./dancer.seen"
03:34:14dongsdont you jsut need cross-gcc and binutils?
03:34:27dongsand, who would want to load apple firmware by default
03:34:48dongsarent you some podcast listener
03:35:07RhapsodyNo, I have some podcasts but I rarely have time to listen to them.
03:37:38 Quit biffhero ("[BX] Tickle-Me Elmo uses BitchX. *giggle* *giggle* *giggle*")
03:38:26 Join biffhero [0] (n=rob@
03:38:43 Quit TeaSea ("Leaving")
03:42:59RhapsodyI get an error on line 8 when I try to patch with your file.
03:44:13 Join BHSPitMonkey [0] (
03:48:27RhapsodyI'll just apply the patch manually.
03:50:13 Quit Daishi ("Client exiting...")
03:52:38 Join jbauman [0] (
03:54:42 Quit hardeep ("[BX] A friend with weed is a friend indeed")
03:54:52GaloisRhapsody: you're right, something changed.
03:55:40GaloisI updated the diff file.
03:55:43RhapsodyIt's okay though, I just read your patch and changed it myself.
03:55:50RhapsodyI have already compiled it.
03:56:20Galoisit's trivial anyway. I don't understand why anybody would want it.
04:12:29RhapsodyHum... I installed it but I can't get it to boot Rockbox, only the iPod firmware.
04:14:41RhapsodyDid I need a new iPod_fw?
04:16:13dongsdoes it show the rockbox boot text?
04:17:46dongswhen you reset
04:17:50dongsyou dont see like 4 lines of crap?
04:17:52dongsok, then you failed
04:18:08RhapsodyJust an Apple for what I believe to be longer than usual.
04:18:23dongsdid you do the steps from step1 for installing stuff?
04:18:29Galoisyou needed to run ipod_fw
04:18:30dongsgetting partition, patchign with fw etc
04:18:48Galoisbasically do everything that you normally do with bootloader.bin except you should use the one you just built instead of the one you downloaded from the rockbox site
04:19:10RhapsodyThat is what I did.
04:19:22RhapsodyI ran ipod_fw, do I need a new copy of ipod_fw?
04:19:32dongsyou dont.
04:19:38RhapsodyI'll try it again.
04:19:39dongsdid you write the patched firmware back to it?
04:19:58RhapsodyYes, otherwise my old Rockbox boot loader should have been launched.
04:24:40RhapsodyRedid it and it works like I wanted.
04:25:05 Quit RoC_MM ("Leaving")
04:25:59 Join qwm [0] (
04:26:09 Join JdGordon_ [0] (
04:27:13 Quit JdGordon_ (Remote closed the connection)
04:28:41RhapsodyIs there someway I can view the .iPod_Control folder from Rockbox?
04:29:21sharpeer... set the file view to all?
04:31:41RhapsodyThis firmware is pretty nice.
04:33:55 Join Jd|uni [0] (
04:34:04 Quit Jd|uni (Client Quit)
04:34:12 Join Jd|uin [0] (
04:34:37Jd|uinany1 around?
04:34:47Jd|uinoh well :(
04:34:55 Nick Jd|uin is now known as Jd|uni (
04:35:44Jd|uniBHSPitMonkey: wanna try out a patch?
04:35:45BHSPitMonkeyJdGordon? that you?
04:35:49 Quit qwx (Read error: 110 (Connection timed out))
04:35:55Jd|uniye, im at uin
04:36:00Jd|unigrr.. uni
04:36:07BHSPitMonkeyerm, I don't really have a build environment set up
04:36:17BHSPitMonkey(and I've never used a patch :()
04:36:37Jd|uniu interested int he jumpt to random dir after the current one is finshed idea?
04:37:04jncwhat about jump to random playlist
04:37:26RhapsodyI'll build one if you like.
04:37:45Jd|unijnc: it could be added...
04:41:23jnci don't own hardware that runs rockbox
04:42:27 Quit MagusG (Connection timed out)
04:42:31 Join george_ [0] (
04:43:01george_quick question, how do i disable the peakmeter since its causing skipping on my 1st gen ipod mini
04:43:31Jd|uniremove the %pb line from the wps
04:43:38Jd|unii _think_ its %pb
04:43:41midkaypm, you mean..
04:43:49george_all right
04:43:50midkayor just load a new one, george_. there are better ones than the default..
04:43:51george_thanks a lot
04:44:00midkayif you wish
04:44:12george_ill look into that. thanks
04:44:22midkay has a couple..
04:44:33 Join MagusG [0] (
04:44:45 Quit george_ (Client Quit)
04:48:35Jd|uniany1 wanna test my patch?
04:48:48jnci would if i had hardware ;)
04:48:52jncthanks though
04:49:17*Jd|uni can compile it for any1 that will test it..
04:53:21dongswhat patch
04:56:04RhapsodyI hope someone merges the SID and MOD player.
04:56:14Jd|uniif u try it.. u need to make 1 change (in the last comment), im heading off now, but lemme know if it actually works if u do try it
04:57:04 Quit Jd|uni ("CGI:IRC")
05:01:32 Join Rob2222 [0] (
05:02:10dongsah thats a useless feature for me
05:05:47 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
05:09:03BHSPitMonkeymidi playback would really please me
05:09:32*BHSPitMonkey hopes for someone to tell him, "you moron, that was perfected 2 weeks ago!"
05:10:23dongstheres patches to midi stuff almost every day.
05:11:33BHSPitMonkeyyeah, I'm semi-aware of that...
05:11:42BHSPitMonkeydoes it work decently on the iPods?
05:11:50BHSPitMonkeys/does it/do they/
05:12:01dongsi havent bothered to try but I could if I had an actual midi file to test with
05:12:41BHSPitMonkeythey're so hard to come by ;)
05:13:05dongswell that and im lazy
05:16:16dongswhats a good gravis patch set?
05:16:48dongsshould i use that 22meg thing off the site?
05:19:50 Quit Rob2222_ (Read error: 110 (Connection timed out))
05:21:48dongson my ipod
05:21:59dongsplaying onestop.mid
05:22:02dongsfrom windows
05:22:08RhapsodyWhat is the key to hold down on the iPod to turn on the backlight in Rockbox?
05:22:17midkaythere is none
05:22:55dongsso midi fails
05:23:04midkay.. so midi is not optimized. right.
05:24:35 Quit amiconn (Read error: 110 (Connection timed out))
05:24:35 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
05:28:09jnci don't know about midi, but xm module playback would be brilliant
05:28:12biffheroI think that 'lock-unlock' will turn on the backlight.
05:30:23biffherowell, that was a guess, really. It does on the retailos, and since rockbox is so amazing, it must happen there, too, right? :-)
05:32:10dongsor touching the scroll wheel
05:32:13dongswill also light up
05:32:19dongsand yes, toggling hold will light up.
05:32:34biffheroI wonder... since it is playing, will 'select' do anything? hit the center select button?
05:32:54dongsone click wont do anthing
05:33:00dongsbut another click on select will send you back to file list.
05:34:11***Saving seen data "./dancer.seen"
05:37:58biffheroso that would work to turn on the light.
05:47:14midkaydongs, the first click not doing anything is a result of "First Keypress Enables Backlight Only" being set to on.
05:47:31midkaypretty pointless on the iPods, since merely touching the wheel will turn on the backlight..
06:05:12 Join Xerion_ [0] (
06:05:49 Quit Xerion (Read error: 104 (Connection reset by peer))
06:05:50 Nick Xerion_ is now known as Xerion (
06:10:15 Quit TCK ("well, if you say so.")
06:30:04 Quit MagusG (Connection timed out)
06:31:28 Join MagusG [0] (
06:36:49 Join gursikh [0] (
06:47:52 Join lBoxy [0] (i=littleBo@
06:51:32 Quit flynux (Remote closed the connection)
06:51:40 Join flynux [0] (
06:54:47 Quit MagusG (Connection timed out)
06:59:44 Quit Rick (Read error: 104 (Connection reset by peer))
06:59:46 Join MagusG [0] (
07:00:04 Quit daurnimator (Read error: 104 (Connection reset by peer))
07:00:14 Join Rick [0] (
07:00:36 Join daurnimator [0] (
07:11:20 Join qwx [0] (
07:18:34JdGordondoes any1 kow how to get rockbox to load a playlist from a file?
07:18:59lBoxyi would like to ask this question: is any chance to put bootloader into Nano from PC/Windows?
07:19:10JdGordonye, check the wiki
07:19:28MikachuJdGordon: 'play' the playlist i think
07:19:38JdGordoni mean in code...
07:19:42lBoxyi use one puter running macosx and another one running win2k and of course windows dont know how to handle MAC formated ipods :-)
07:20:04midkayJdGordon, did you try checking tree.c or something for like TREE_ATTR_M3U?
07:20:19JdGordonno.. not yet :p
07:20:21midkaysee what it does with the file.. probably something like playlist_load(). (sounds familiar to me)
07:20:23lBoxymac reads Windows formated iPod fine
07:21:31 Quit qwm (Read error: 110 (Connection timed out))
07:27:10 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:29:45 Nick qwx is now known as qwm (
07:34:15***Saving seen data "./dancer.seen"
07:35:27JdGordonwhat playlist file extensions does rb support? just .m3u?
07:41:57JdGordonjust checking... playlist_create() can load a .m3u file as the new playlist?
07:46:05 Join Acksaw [0] (
07:46:22 Quit gursikh ()
07:47:01 Join nave7693 [0] (
07:47:38 Nick slimeball is now known as _slimeball (i=jew@
07:47:42 Part nave7693
08:06:36 Join Sinbios [0] (
08:21:52 Quit aegray (Connection timed out)
08:22:23 Join aegray [0] (n=aegray@
08:25:29 Quit midkay ("Leaving")
08:25:31 Join midkay [0] (n=midkay@rockbox/developer/midkay)
08:39:15 Join bytie [0] (n=zic@
08:39:58 Quit bytie (Remote closed the connection)
08:53:07amiconnpreglow, whatboutbob: Here's a new patch to test:">
08:54:04amiconnIt recovers from DMA error, and also discards DMA blocks where S/PDIF errors happened (counter not advancing with bad signal or no signal)
08:54:21 Join petur [0] (
09:01:49whatboutbob_amiconn: thanks. will start testing in about an hour (just heading home from work now)
09:02:09peturheh.. I just arrived at work :D
09:04:40 Join RedBreva [0] (
09:06:30whatboutbob_amiconn: so the best way to test the changes would be to mess with the cable/connector?
09:07:50*whatboutbob_ points at petur and laughs in a high-pitched 'Haha', Nelson-style
09:08:44*petur is glad his pc doesn't produce sound
09:10:03 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
09:10:34whatboutbob_If this optical recording patch does iron out the bugs, is there any chance it would be committed for 3.0...or is it considered 'new' functionality, therefore out of scope?
09:11:40 Quit MagusG (Connection timed out)
09:16:33 Join MagusG [0] (
09:18:00linuxstbwhatboutbob: I don't know what others think,. but if it's relatively bug-free, I think it should be in 3.0 - it comes under "core functionality", rather than as a new feature.
09:22:22amiconnI would even call it a bug fix (the recovery from DMA error surely is)
09:23:18amiconnThe recovery from stopped DMA still isn't bulletproof (namely an overrun due to inability to save in time, e.g. harddisk shaking) will cause the same symptoms
09:24:05amiconnThe dma/spdif error counter should be displayed somewhere, to make the user aware errors occured, and blocks dropped
09:25:26 Join Paul_The_Nerd [0] (
09:27:17 Join B4gder [0] (
09:27:17amiconnSome ifdefing is also missing in the patch (code parts that are irrelevant for H300 and X5)
09:29:04amiconnIt's mainly preglow's patch with an ifdef typo corrected and the DMA recovery + s/pdif error check
09:31:30 Join shrizza [0] (
09:31:56shrizzaSo uh, rockbox doesn't support Firewire?
09:34:18***Saving seen data "./dancer.seen"
09:34:26 Join matsl_ [0] (
09:34:32amiconnmidkay: I wonder what your sokoban fix should fix, resp. why it's necessary
09:34:48amiconnColours should be reset to standard anyway by the plugin loader
09:35:27shrizzaAlso, how can you force rockbox to shut off.
09:35:39shrizza(on an iPod)
09:36:42shrizzaOh, nevermind on that question.
09:36:48shrizzaThough I'm still wondering about Firewire.
09:37:08Paul_The_NerdRockbox doesn't really support USB either, but we don't automatically reboot you for firewire connections.
09:37:34whatboutbob_amiconn: do you know if the DMA error is caused only by a faulty signal? ie there is nothing 'wrong' with the patch that causes the errors (previously hangs)...and now rockbox just handles the faulty signal better?
09:37:58 Join ender` [0] (i=ychat@
09:38:05amiconnIt seems so. I didn't change the way how DMA is set up
09:38:29amiconn(though I still wonder why a faulty spdif signal can cause a DMA 'configuration error')
09:38:37amiconnMust be a glitch in the coldfire
09:38:46shrizzaRight now, in order to connect my iPod to my computer, I shut off rockbox and boot back into the original Apple firmware, then connect the firewire. I'm just wondering if this is the best way to connect the iPod to the computer.
09:39:04shrizzaOr if there's a faster way to achieve this without having to switch firmwares.
09:39:18Paul_The_NerdWell, when you plug in a USB cable, it shuts down Rockbox and reboots into the Disk Mode anyway.
09:39:30Paul_The_NerdA little bit faster, but not much.
09:39:53shrizzaI've never tried USB and don't plan on it because my computer doesn't support USB2.
09:40:02amiconnLinusN: Is there a reason why you reserved vectors for the interrupt controller 2 in the vector table, but never set INTBASE nor SPURVEC? Interrupt controller 2 is totally uninitialised...
09:40:15Paul_The_Nerdlinuxstb: Are you around by chance?
09:40:25LinusNamiconn: of course there is a reason
09:40:40shrizzaOh, I'm also wondering about the sensitivity of the click-wheel for activating the backlight. It's awfully sensitive isn't it?
09:40:50B4gderyes, but is there another reason that fascism? ;-P
09:40:52amiconnAlso, it seems you used 'reserved' areas for a couple of interrupt vectors
09:40:59*B4gder stops trying
09:41:23LinusNthe reason is that i started with the vector table, and didn't find a use for any of the interrupts
09:41:33 Quit matsl_ ("Leaving")
09:41:40amiconnWell, we will need the GPIO interrupts for usbotg
09:41:42LinusNso i never got to the initialization part
09:42:00LinusNdid i use reserved vectors?
09:42:55amiconnYes, you 'foled' a couple of vectors in the space reserved for fp exceptions
09:43:03 Join zigford [0] (
09:43:15amiconnDunno if that matters...
09:43:31 Join Kohlrabi [0] (
09:43:53amiconnThe i2c interrupts might also be useful. i2c2 is on interrupt controller 2
09:43:57linuxstbPaul_The_Nerd: Yes, for a few minutes.
09:44:56Paul_The_Nerdlinuxstb: I was just gonna say, the USB resetting continually problem has shown up on a user's Nano. And if you extract Bootpartition.bin and then write it back, it has the problem, but if you restore the iPod it doesn't, so it sounds like it's a ipodpatcher.exe bug.
09:45:39linuxstbYes, I read that.
09:45:56Paul_The_NerdActually, did they say they use iPodPatcher.exe? I just realized, I don't know which method they used.
09:47:17 Quit lostnihilist (Remote closed the connection)
09:48:36linuxstbRestoring the ipod will probably mean that the flash bootloader gets reflashed. That could be the difference.
09:49:08 Join piroko [0] (
09:49:41Paul_The_NerdBut shouldn't an ipod that you've done ipodpatcher -r N blah.bin then ipodpatcher -w N blah.bin end up in the exact same state as where you started?
09:50:01linuxstbYes. So I'm sceptical that the user correctly restored an unmodified bootpartition.bin.
09:50:18 Join damaki_ [0] (
09:50:38qwmthe "features we will not implent" list is empty?
09:51:12 Part aziiz
09:51:24linuxstbqwm: Where?
09:51:44qwmis it available elsewhere?
09:51:54qwmwhere does it come from?
09:52:45linuxstbIt's here:
09:52:54qwmthanks. :)
09:53:22linuxstbBut most of them apply to the Archos line of players.
09:54:24Paul_The_NerdSome of the reasons are perhaps a bit outdated. The ROM can only access FAT32 for example.
09:54:46qwm"EXCEPTION: Certain newer players (i.e., the international version of the iriver H300) are capable of acting as a master." - what about the ipods?
09:55:14linuxstbqwm: It's a wiki...
09:56:01qwmi meant that as a question, i don't know whether the ipods are capable of it or not. :p
09:57:36linuxstbYes, it's technically possible. But extremely hard without PortalPlayer datasheets.
09:57:46qwmi see. :\
09:59:41 Join lostnihilist [0] (
10:00:13linuxstbPaul_The_Nerd: IMO, we should remove the alternative filesystems support from the NoDo. Even though no-one thinks it's worthwhile, it is technically possible.
10:00:17 Join _FireFly_ [0] (
10:01:11Paul_The_NerdWith the new MAS codec thing, can it record to wav now?
10:05:02 Quit MagusG (Connection timed out)
10:05:22linuxstbNo, that hasn't been implemented yet.
10:05:41linuxstbBut yes, it's possible.
10:06:36 Quit damaki (Read error: 110 (Connection timed out))
10:06:47Paul_The_NerdI'll update the NoDo a wee bit then.
10:09:25 Join MagusG [0] (
10:14:15dongswhy the hell would you want "alternative filesystems"
10:15:17B4gderlarger files than 4GB
10:17:09LinusNhis reasons are a bit vague imho
10:17:22LinusN"wish I could use it as backup media for images"
10:17:24crwlsmaller cluster size for the high voltage sid collection? ;)
10:17:40LinusN"I certainly will not cripple my good named files"
10:17:51B4gderyeah, those are not file system limits
10:18:04LinusNB4gder: yes they are
10:18:15KohlrabiDoes anyone know somethign about CUE-support in rockbox?
10:18:16B4gderaren't most of them windows limits?
10:18:21Paul_The_NerdKohlrabi: It does not exist.
10:18:25LinusN4gb image files and fat32 file name limits
10:18:40KohlrabiOr unofficial patches?
10:19:00Paul_The_NerdKohlrabi: It wouldn't work well with the playlisting method currently used, and nobody's actually named a real benefit. And the people who want it haven't made any patches, I believe.
10:19:01LinusNKohlrabi: no real plans
10:19:14KohlrabiSince f2k dumped mutifile-CUE-support I have to encode to single file :\
10:19:23Kohlrabiaw, too bad
10:19:42Mikachurockbox plays losslessly with multiple files too
10:19:47LinusNwhy single file?
10:20:18Kohlrabiback then I used EAC to rip and encode to mutiple files without gaps, and had the gaps appended to previous track
10:20:27KohlrabiWhich is stated noncompliant
10:20:30LinusNso split the file then
10:20:33KohlrabiBut still worked in f2k0.8.3
10:21:34Paul_The_NerdWith proper formats there's not much use for a CUE file relating to playback.
10:21:42LinusNthe problem is that cue file support is really hard to fit into the playlist structure of rockbox
10:22:00LinusNit requires major changes to work well
10:22:09B4gdercue files are a hack anyway
10:22:11*B4gder ducks
10:22:11Kohlrabiyeah, I understand that, jsut asking
10:22:17LinusNi know
10:23:44linuxstbBut there are more reasons than just ".cue" to break Rockbox's "one file = one track" - the tracker formats have multiple tracks in one file, we can have chained Ogg files, and chapter marks in .mp4 files (I'm guessing podcasts and audiobooks use them).
10:25:51Paul_The_NerdWhat exactly is the big conflict with the current playlisting system?
10:27:02linuxstb1) Playlists only specify a file, not a track within a file.
10:27:37Paul_The_NerdLike, if you didn't allow reordering of the internal tracks (and just treated multi-track files as one long single track in a normal playlist, or as a playlist of their own if opened individually so a user can actually browse chapters) would that be doable?
10:27:39Mikachufor files with subtracks that may have to change anyway... sid and nsf etc
10:27:45linuxstb2) The get_metadata() function would need to return a list of metadata structures (one per track) instead of just a single metadata structure.
10:28:59LinusNthere is a patch from the stoneage which implements cue file support
10:29:02linuxstbNot related to the playlists problem, but we can't accurately seek in VBR MP3 files - meaning .cue support for MP3 will never be perfect (unless we frame-walk the entire file which will be very slow).
10:29:20linuxstbLinusN: Yes, I did that... But you're right, it's from the stoneage.
10:35:12 Quit jbauman (Read error: 110 (Connection timed out))
10:36:03 Quit Sinbios ("If the definition of a klutz is someone who doesn't have eyes on their ass, then yes, I suppose I am a klutz.")
10:37:27 Join Sinbios [0] (
10:42:02dongsrockbox works good on a mini, right?
10:42:07dongsno skips on mp3s?
10:42:21Mikachui think it works well
10:42:34Mikachushould be the same as other ipods
10:42:53Paul_The_NerdThe mini (and grayscale 4gs) I believe work a bit better
10:43:13Paul_The_NerdThough I didn't have MP3 skips *before* the changes to the MP3 codec, on my Nano.
10:43:33Mikachui think the issue is only if you use the eq at the same time
10:44:09Mikachuand i don't think any of the ipods handle that well
10:45:10Paul_The_NerdEQ. :-P
10:45:23Mikachui don't have many mp3s so i don't know for sure :)
10:45:39*whatboutbob kicks himself
10:45:46whatboutbobstupid optical out...
10:49:03daurnimatorhi all
10:49:21daurnimatori'm looking for a tiny mp3 player with a huge hdd
10:49:55B4gderwhat's huge to you?
10:50:30*whatboutbob fights the urge to say 'oh, about 8 inches'...
10:50:50Paul_The_NerdSo, the X5 60gb.
10:50:57Paul_The_NerdOr an iPod 5G 60gb
10:50:58linuxstbThe 60GB ipod video/5g...
10:50:58daurnimatorif it has movie playing ability, then 80gb+
10:51:49daurnimatoralso something with a nice UI
10:51:57Paul_The_NerdWrong place to ask then.
10:52:01B4gderhey, this is #rockbox
10:52:13Mikachurockbox doesn't have a nice ui?
10:52:17scorchelinuxstb: speaking of which, are you gonna do something with that portable guy;s player?
10:52:29daurnimatorUI meaning button layout
10:52:40Mikachuthat's a new meaning
10:52:47B4gderdaurnimator: and what is "nice" button layout to you?
10:52:52linuxstbscorche: I'm waiting to see what documentation and sourcecode is forthcoming from the manufacturer.
10:52:53daurnimatorUI = user interface
10:53:08daurnimatoruser interface = ways you interface = buttons = layout
10:53:17linuxstbscorche: I'm definitely interested in it - but not yet committed.
10:53:41B4gderifaik, there are only 60GB available in players with 1.8" disks and there are two that run Rockbox
10:53:41scorchelinuxstb: lemme know...something new + possible video + 2.5 hd = win in my book
10:54:08linuxstbscorche: Are you interested as a developer, or as a user?
10:54:13daurnimatoroh, btw, any1 know where i can get an MK4007GAL?
10:54:16Paul_The_NerdYeah, if Rockbox moves in on that player, I'm probably gonna hafta scrape together some cash.
10:54:35 Join mikearthur [0] (
10:54:35scorcheboth, however, at the moment, i have no time to dedicate to anything of the sort
10:54:47ravonBest christmas album ever.
10:54:56daurnimator(MK4007GAL = biggest single platter 1.8" drive) - made by toshiba
10:55:17B4gderyou're now talking _single_ platter player?
10:55:27amiconnImho subsong support (cuesheet, tracker formats, ogg) would first require to drop the current playlist format
10:55:27daurnimator[18:53:47] <daurnimator> oh, btw, any1 know where i can get an MK4007GAL?
10:55:44B4gderyes but you asked for players with huge disks
10:55:49daurnimatori know
10:56:04amiconn...which is bad, imho
10:56:07daurnimatorbut my current one has a low profile 1,8" 20gb drive
10:56:13daurnimatorand i could just upgrade that
10:56:14amiconnm3u is nice and simple, and widespread
10:56:37B4gderI don't see why we would need to drop it
10:56:46Mikachuyou can have # comments in .m3u
10:57:24whatboutbobamiconn: nice work w/ spdif.
10:57:28Mikachujust make a #ROCKBOX:stuff like winamp(?) added #EXTINF:title
10:57:43amiconnYeah, but comments are comments, and if you process this m3u with a not extension-aware program, the comments will probably be dropped
10:58:00Mikachuif you changed the format you couldn't use any other program at all
10:58:29amiconnThere are more flexible playlist formats, but they're not as widespread
10:58:50B4gderenter RPF - Rockbox Playlist Format
10:58:54*B4gder runs
10:58:54markundaurnimator: I only know some places where you can buy the 30GB ones
10:58:57whatboutbobamiconn: i'm giving myself a headache listening to chopped up songs tho...
10:59:04daurnimatormarkun: link?
10:59:15daurnimatormaybe i can lead on from them
10:59:35Paul_The_Nerdwhatboutbob: Chopped up songs?
11:00:17amiconnB4gder: The Deliplayer playlist format can handle subsongs, sub-lists etc
11:00:56whatboutbobpaul_the_nerd: testing spdif a song, record it, pull plug out, put it back in etc...ends up one screwed up recording...but recording doesn't hang. :)
11:02:36_FireFly_ whatboutbob: why should recording hang if you pull the plug ?
11:03:08whatboutbob_firefly_: with the initial patch they were hanging...seems to be fixed now (bravo amiconn!).
11:07:52 Join webguest73 [0] (
11:08:27 Quit _FireFly_ ("Leaving")
11:12:42 Quit Poka64 ("XChat 2.6.2 -, XC-WSys 0.18")
11:13:53preglowamiconn: roxor!
11:14:47 Quit MagusG (Connection timed out)
11:15:26whatboutbobpreglow: seconded. :)
11:15:56*preglow summons mborus
11:16:40whatboutbob...and his faulty connections. :-)
11:18:09 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
11:18:25preglowi also want to find out if my new sample rate detection is really necessary
11:19:08 Join dwihno [0] (n=dw@
11:19:31 Join MagusG [0] (
11:22:48 Quit phaedrus961 (Read error: 110 (Connection timed out))
11:23:53*whatboutbob wishes he could get his puter's optical out working...
11:24:12amiconnpreglow: The sample rate isn't displayed correctly, but it works for saving the wav
11:24:24*amiconn tried recording from dvd
11:24:44whatboutbobdoes it save different sample-rate headers correctly?
11:25:01preglowamiconn: mborus has a source that doesn't report the correct sample rate
11:25:39preglowthe sample rate displayed is not right, no
11:25:46preglowcan fix that
11:25:57 Join phaedrus961 [0] (n=dfdsfsdf@
11:26:06whatboutbobeither of you two mind if i post a pre−−compiled build w/ the new patch to the forum?
11:26:16preglowgo ahead
11:26:24preglowjust make it clear what changes amiconn applied
11:29:21 Join Pi [0] (
11:32:31 Quit RedBreva (Read error: 110 (Connection timed out))
11:32:58whatboutbobpreglow: err...i did my muddled best...
11:34:20***Saving seen data "./dancer.seen"
11:35:56JdGordonany1 know anything about data-flow-diagrams ?
11:36:50 Join Cassandra [0] (
11:37:26scorcheJdGordon: idont think anyone knows "anything" about them
11:37:41JdGordon:'( im gonna fail then more specific
11:38:07JdGordonna its all good.. good ol google saves the day!
11:41:17 Quit scott666 (Read error: 110 (Connection timed out))
11:43:44 Quit mikearthur ("Konversation terminated!")
11:46:12CassandraMy brain is stuck in C++ mode. I've lost the ability to write perl!
11:46:32scorche<3 python
11:46:38*JdGordon beats Cassandra with the perl stick!
11:46:46JdGordonhope that helps...
11:47:13CassandraThe beatings will continue until morale improves.
11:47:34CassandraPython looks good, I just can't be arsed to learn *another* scripting language.
11:47:40CassandraIt's too confusing as it is.
11:47:49scorchepython is simple
11:47:50*JdGordon beats the whole channel... "You will be cheery damn you"!!
11:48:03scorchewhich is one of the things that makes it so great
11:48:32*scorche is just getting into wxpython as well
11:49:11scorcheand on that note, i am out...hopefully, i can get at least 2 hours of sleep tonight..
11:53:17CassandraRemind me how to make .* non-greedy in perl regexps, would you?
11:53:46Cassandra2 hours - what a lightweight. ;)
11:54:24CassandraAh .*?
11:55:13preglowpython is crap
11:55:15preglowgod, i hates it
11:55:38preglownon-greedy is ? after the expression, yes
11:55:43scorchepreglow: to each his own
11:55:48JdGordonis ftp faster than scp ?
11:56:09preglowJdGordon: well, at least it doesn't waste time on encryption
11:56:13 Quit dwihno ("trasa :(")
11:59:07 Join dwihno [0] (n=dw@
12:00:40daurnimatorhi all
12:00:48daurnimatorhow does one make an audio codec?
12:02:46CassandraOne gets lostlogic to do it for you. :)
12:03:00CassandraOr someone else who understands codec magic.
12:03:09CassandraFailing that, RTFS.
12:03:50Cassandrapreglow: Should be. It's still being greedier than I want it to be though for some reason.
12:04:38 Join Moos [0] (
12:05:15CassandraEither there's a bug in perl, or I'm subtly failing to understand regexp syntax.
12:05:22CassandraGuess which my money is on.
12:05:27preglowdaurnimator: you make a codec plugin, there are a tons of them in apps/codecs
12:05:37preglowdaurnimator: usually, you just wrap a codec library you've imported from somewhere else
12:06:02preglowregex flow = magic
12:09:47daurnimatori'm going to try and make my own wav interpreter from scrath
12:09:49daurnimator - in lua
12:10:54CassandraHah! worked around it. Take that, interpreted language of Satan!
12:12:39whatboutbobpreglow/amiconn: the patch is looking pretty good. I'm just analysing a recording now that i pfaffed about with for an hour.
12:12:49Mikachui thought '?' is 0 or 1 matches
12:13:40CassandraMikachu, a lot of the character meanings in regexps are overloaded depending on context. (cf ^)
12:14:26CassandraPersonally I think it's kind of sloppy and creates extra confusion, but they never bothered asking me when they were designing them.
12:16:10whatboutbobpreglow/amiconn: any chance either of you could auto-enable spdif out when entering digital recording screen?
12:16:46Mikachuhm, i guess so
12:16:57Mikachuregex is a write-only language
12:17:05Mikachuat least, i can't read it
12:17:17CassandraVery few people can.
12:17:38CassandraIt takes the kind of mind that can parse or win the obfuscated C contest.
12:17:46*Cassandra knows someone who won that.
12:18:09CassandraHe's very clever. Works for the University of Cambridge Computer Service.
12:19:56*amiconn spots Cassandra
12:20:28amiconnCassandra: Any news regarding at&t?
12:21:31JdGordonany1 know how to choose which users are allowed to loginto ftp on linux? im using the proftd server
12:22:40Cassandraamiconn: Only that I have been so busy with my new voluntary work that it's completely slipped my mind.
12:22:53CassandraI'll try to get onto it in the next day or two.
12:23:35Cassandraftp? That's so 20th Century. ;)
12:24:36CassandraIt's been so long since I used it I can't remember
12:26:26preglowwhatboutbob: i will
12:26:56whatboutbobpreglow: gracias.
12:29:46whatboutbobpreglow: the source is still occasionally defaulting back to mic.
12:30:48*JdGordon is an idiot... it was working the whole time.. i was using the wrong bloody password :'(
12:33:50 Quit aegray (Read error: 110 (Connection timed out))
12:39:24 Part Paul_The_Nerd
12:40:37MikachuJdGordon: users listed in /etc/ftpusers aren't allowed, everyone else is
12:40:48MikachuJdGordon: you can use a virtual passwd for proftpd if you want though
12:41:55 Join TeaSea [0] (
12:42:02whatboutbobpreglow/amiconn: Congrats (and thanks) again guys. I can't fault spdif recording at 16/44.1. Will try again tomorrow. I'm off to bed.
12:45:44 Nick _slimeball is now known as slimeball (i=jew@
12:45:59*whatboutbob is curious to see what happens with the 2GB autosplit when the 'time' is off ("incorrect") due to signal errors.
12:47:24preglowwhatboutbob: yeah, i don't know why it does that
12:49:06 Join Poka64 [0] (
12:49:16 Quit gtkspert (Read error: 104 (Connection reset by peer))
12:51:26preglowamiconn: do you know if the control channel data is something the app usually pieces together itself, or is it something the spdif interface itself usually constructs?
12:51:43preglowamiconn: i'm wondering why mborus got a wrong sample rate from his satellite receiver
12:52:27 Join gtkspert [0] (
12:52:31 Join ashridah [0] (
12:52:46 Join mikearthur [0] (
12:54:57 Join akaidiot [0] (
12:56:43whatboutbobdoes anyone know off the top of their head how often (or, er, when) the recording buffer writes?
12:56:48amiconnpreglow: :shrug: Do you mean for input or output?
12:57:11amiconnwhatboutbob: It writes whenever the buffer is almost full
12:57:36whatboutbobis the buffer a set size?
12:58:03amiconnIt's the audio buffer
12:58:30whatboutbobk. thanks.
13:01:22preglowamiconn: input
13:02:03preglowamiconn: in the patch you got, i just find the sample rate from the control channel register, and it's sometimes wrong, it seems
13:03:34amiconnWhen do you read the register?
13:04:14amiconnIt worked for me, but then I only have one source, and can only test 44.1kHz (CD) and 48kHz (DVD)
13:04:38preglowi just ripped that code from the debug screen analyzer, i don't have the specs for it
13:04:45preglowbut linus got the values from a spec
13:05:04preglowhowever, i've got some newer code that uses direct clock measurement
13:05:10preglowwhich should work in all cases
13:05:15amiconnPerhaps I should get the optical sender chip as well, then I can convert coax->optical and test 32kHz
13:05:28amiconn(by playing a 32kHz mp3 on my archos)
13:06:47preglowi think the optical out on my chipset is locked to 48khz
13:06:52preglowbut i could boot windows to check
13:07:05 Join dpro [0] (
13:10:21*Cassandra discovers bugs in the WPSes. Oh joy!
13:12:54 Quit solexx (Read error: 104 (Connection reset by peer))
13:12:55 Join solexx [0] (
13:13:24 Quit akaidiota (Read error: 110 (Connection timed out))
13:13:50ashridahCassandra: what, only one?
13:14:01ashridahoh. bugs plural
13:15:29preglowlostlogic: present?
13:16:02 Join RedBreva [0] (
13:34:24***Saving seen data "./dancer.seen"
13:34:50 Quit MagusG (Connection timed out)
13:35:28 Join MagusG [0] (
13:35:45*petur thinks ipod users can't read
13:36:07peturthere must be a ton of identical bugreports about skipping on ipod
13:36:51peturshould we keep the oldest report and close the others as IDENTICAL?
13:36:56RedBrevaI think iPod users should just use it properly and then it dosn't skip!
13:37:14B4gderpetur: or keep a newer if there happens to be any particular with more info than the others
13:39:08dwihnothat reminds me
13:39:17dwihnothe ipod port has skipping problems ;)
13:39:24 Quit mikearthur (Remote closed the connection)
13:39:44*petur slaps dwihno with a large bugreport
13:39:47 Join mikearthur [0] (
13:39:56CassandraI have decided I hate WPSes. Let's go back to all text.
13:40:09dwihnoI 2nd that!
13:40:10preglowwps editor!
13:40:20B4gdertext is good for you
13:40:52CassandraYou wouldn't believe how difficult a coding job it is just getting the right size WPSes in the right ZIPs.
13:41:26dongsdoesnt drawing all that shit on a wps waste cpu time
13:41:32dongssolid background and text only plz
13:41:52dwihnoTons of animated stuff! Like in the early web years!
13:41:53CassandraI'd have thought you'd at least have some sympathy, Bagder. After all, it's your mess I'm cleaning up. (Joke.)
13:42:24B4gderhey, my mess at least did half the job! ;-)
13:43:10B4gderit would be the upper half since I always mess up
13:43:29lostlogicpreglow: *yawn*
13:43:57peturit's a pitty you can't merge tracker entries
13:44:01preglowlostlogic: so, how long time do i have to supply new data when pcm_play_data calls me back?
13:44:27B4gderpetur: yeah, that woudl've been handy
13:44:38lostlogicyou're called in the ISR, so not long at all...
13:44:47lostlogicbut I don't actually have numbers on that.
13:45:02LinusNit depends on the fifo setting
13:45:15lostlogicLinusN: in the DAC?
13:45:15preglowlostlogic: ahh, right, so if i am to actually generate data, the wisest is to generate it outside the callback
13:45:23lostlogicpreglow: yeah
13:45:28preglowthat explains why i get crackling all over the place
13:45:35preglowbut i still get some when i do it the other way
13:45:38preglowjust very, very occasionally
13:45:40LinusNlostlogic: the coldfire fifo
13:46:20lostlogicLinusN: which is where it the DMA pulls from to send to the DAC?
13:46:42LinusNthe dma pulls from memory to the fifo
13:46:43preglowsends to
13:46:49preglowthe fifo is six deep, no?
13:46:57preglowsix samples isn't really much time
13:47:06preglowapparently not at any rate
13:47:06LinusNbut i believe you can set the threshold for the interrupt
13:47:12preglowi'll need to use a very low block size
13:47:26lostlogicyay DMA interrupt storms :)
13:50:59preglowpcm_play_data isn't in the sims ://///
13:51:04lostlogicpreglow: what are you working on any way? It can't be pcm buffered?
13:51:26preglowjust doing some small fiddling with a sine sweep generator
13:51:45preglowit's supposed to have a frequency reading, so it can't be buffered
13:51:52preglowthe reading would be a bit hard to predict, in that case
13:52:09 Join PaulJam [0] (
13:52:10 Quit Sinbios ("If the definition of a klutz is someone who doesn't have eyes on their ass, then yes, I suppose I am a klutz.")
13:52:40preglowi wonder why i still get some clicking
13:53:02preglowthe only thing the callback does now is set the address and length of the buffer, then schedule another synth round
13:53:28preglowdoes pcm_play_data copy the buffer?
13:53:38lostlogicit plays it in place
13:53:39preglowthat'd explain it, then
13:53:51preglowwell, i should have ample time for memcpy
13:54:00preglowi could also do double buffering
13:54:24LinusNlooks like the programmable threshold is only for recording
13:54:49LinusNthat means that you have less than a sample period to handle the dma irq
13:54:59preglowhaha, ouch
13:55:02preglowdouble buffering it is, then
13:57:26preglowand another fun problem: adequate frequency accuracy for long sweeps
13:57:47 Quit piroko ("telinit 0")
13:57:55*preglow misses floating point
13:58:01*Cassandra beats herself with a stick/
13:58:20CassandraI just spent the last half an hour chansing a bug that was the code functioning correctly.
13:58:32lostlogicat least you don't hafta fix it now :)
13:59:34preglowlow latency audio in a rockbox plugin is no fun
13:59:44preglowyielding is basically out of the question
14:00:14Cassandrapreglow: Better implement pre-emptive multi-tasking then. ;)
14:00:15lostlogicpreglow: why was using the pcmbuf_get_latency (or wahtever itis) to determine the currently playing freq?
14:00:29lostlogicthat should get you very close to accurate latency numbers
14:00:45pregloweven when the buffer isn't entirely filled?
14:01:01lostlogicit measures the unplayed bytes + the number of bytes waiting on the DMA or fifo
14:01:11preglowohh, that might be good then
14:01:37preglowneed to stuff a host of new functions in the plugin api, then
14:01:56lostlogicgee fun :(
14:02:22preglowbut it would be very convenient
14:02:34LinusNpreglow: doing a frequency sweep?
14:02:51preglowLinusN: yeah, febs wanted one, and i figured it'd be a quick deal...
14:03:27preglowworking fine, the sweep is just a bit inaccurate
14:04:04preglowdelta = (endfreq - startfreq)/(duration*SAMPLE_RATE) <- and god knows why...
14:04:52*amiconn doesn't think the frequency sweep is 3.0 stuff...
14:07:16LinusNpreglow: looks like that formula would give very small deltas
14:07:22preglowLinusN: exactly
14:07:30LinusNno wonder it's inaccurate then
14:07:35preglowi need to do some fixed point stuff
14:07:37preglowi can't just cut of the fraction
14:07:43preglowit's important when the sweep is thirty seconds
14:08:59LinusNyeah, just shift the delta <<16 or something
14:09:12LinusNor rather (endfreq-startfreq)<<16
14:09:26*amiconn would think a large-range sweep should be done exponential
14:09:41 Quit MagusG (Connection timed out)
14:09:47dpropreglow: why not use a table a la csound ?
14:09:51preglowamiconn: it should, i'm just doing a linear one to verify all else works
14:09:55preglowdpro: i do, of course
14:10:11preglowbut that table still needs fractional indexing
14:10:27preglowwith the table index delta growing with time
14:10:31dpropreglow: how big a table are you using ?
14:10:38preglowdpro: 128 samples with linear interpolation
14:10:54dpropreglow: ic why not simply make it bigger ?
14:11:00preglowbecause it doesn't alter the problem
14:11:04preglowand smaller tables are better
14:11:28 Join MagusG [0] (
14:11:38preglow'make the table bigger' sounds like a very csound-ish solution to any problem, heh
14:12:01preglowthat and 'make krate=arate' are the two i've heaC[C[Crd the most, no doubt
14:12:11dpropreglow: because it helps at higher frequencies if all you have is a tabosc
14:12:22preglowdpro: it's a sine wave, the is no freq attenuation
14:13:20*dpro will go to the roof now and see if wifi is still working there
14:13:47preglow128 samples is plenty for a sine anyway, the first harmonic is at -80 db
14:14:16 Join solexx_ [0] (
14:16:09dproyippie ... link level sucks but is way enough for ssh and irc ;)
14:16:53*dpro goes to look for sunblocker ;)
14:19:00dprodamn I still need a damn 1.0 retail firmware for 5g ... BTW the instructions on the wiki on how to find a bootpartition.bin on the cd/ in the ipod updater mpkg on os x are simply wrong ... there are no firmware-x.x.x files neither on the cd nor in the package
14:20:42dongssure there are
14:20:45dongsthey're in teh resource fork
14:20:54dongsor whatever
14:23:56dprodongs: believe me I tried on os x and linux there is nothing but manuals in content/resources
14:24:16dongshave you tried in: WINDOWS XP PROFESSIONAL SP2?
14:24:41dprodongs: no I don't have win
14:25:52dprodongs: in fact I have the install cd mounted (as hfsplus) right now on my linux box and there is _no_ firmware file
14:28:57dongsRight click the "iPod Updater" (or later one), click "Show Package Contents" and navigate to "Contents/Resources/Updates/". Copy out the firmware partition image that match your iPod, as shown in the table below.
14:29:02dongstheres nothing like this?
14:29:52dprono there is no Updates folder in Contents/Resources neither in the 2006-10-1 nor on my cd nor in the 2005-sth updater -> instructions are wrong
14:30:02 Quit solexx (Read error: 110 (Connection timed out))
14:33:27 Quit mikearthur (Remote closed the connection)
14:36:44PaulJamIs it possible to add a comment to a Flyspray entry annonymously?
14:42:58JdGordondoes rb have strstr?
14:44:23B4gderI think so
14:44:47JdGordonim getting undefined reference to strstr when i do a target build :'(
14:47:02B4gderfind and grep are your friends
14:47:17JdGordonwe have strcasestr not strstr which is close enough...
14:47:19B4gderit seems we only have a strcasestr() which is a case insensitive version of it
14:52:01JdGordonPaulJam: wanna try out a new version of the patch?
14:52:49 Join cv [0] (
14:52:53PaulJamJdGordon: sure, i had already done the cahnge ti the patchfile and it seems to work
14:53:19cvHi...just a stupid question: where can i download the plugins listed in the rockbox-wiki? i can't see any link?!
14:53:41JdGordonye, i didnt like part of that patch so ive reworked it a bit and it can (hopefully) load a random playlist instead of a folder..
14:54:10dongscv: which plugins?
14:54:10 Join damaki__ [0] (
14:54:13dongsthey're all included.
14:54:17dongslook in menu/browse plugins.
14:54:25PaulJamdoes that mean i have to create a playlist for every folder?
14:55:04cvdongs: For example dictionary or Splitedit, or doom?
14:55:11dongsyes doom is there
14:55:14dongsdict too
14:55:18dongsnot sure about splitedit.
14:55:19JdGordonno, u can set it to load a random msu or folder, or only a random folder, or only random msu, or only the actual next folder.. works on your current files...
14:55:54dongscv: press 'menu" button until you see a menu like blahblah settings, and 2nd from bototm entry is "browse plugins"
14:55:58cvdongs: oh you're right...but doom isnt included
14:56:04dongsthen its not on your platform
14:56:07dongswhat are you on?
14:56:29dongsthat might be why. did you see a checkmark next to doom plugin as being available on it?
14:57:16 Quit Cassandra (Read error: 113 (No route to host))
14:57:44PaulJamJdGordon: would it be possible to make it also work with the butoncombo for next directory?
14:57:46cvdongs: *g* i knew im just too stupid ;) thanks anyway :)
14:58:05dongsi think daily builds per each target only include sstuff known to work for that particular target.
14:58:14JdGordonPaulJam: i tihnk in the reworking ive done it so it does... but cant test it on my player so i dunno
14:58:14dongsso that explains why you have no doom
14:58:22markundoom is included in the H120 builds
14:58:25cvdongs: okay...ive have doom, doom is included ;)
14:58:52cvdongs: but i didnt' see it ;) splitedit isnt included
14:59:35dongsthat i duno, ive never heard of it.
14:59:39 Join KN|stiff [0] (i=phhome@
15:00:07Moossplitedit wasn't adaptated for swcodec platforms yet, just working for archoses
15:00:25Piaccording to the wiki, split edit is only for the Recorder & Ondio
15:02:16JdGordon:'( i tinhk the usb port on m h300 is almost dead
15:02:22 Quit zigford (Read error: 113 (No route to host))
15:02:22cvok, question again: i heard from an autoupdater? is there any and when yes, where can i find it?
15:02:56Mooscheck misticriver forums and sure you'll find out
15:03:37 Quit ashridah ("Leaving")
15:05:35 Nick slimeball is now known as _slimeball (i=jew@
15:05:52dprocv: you'll need a .wad file to play doom of course
15:06:15cvdpro: ok, i read this, but where to get this wad-file?
15:07:38 Join akaidiota [0] (
15:09:11 Join Jungti1234 [0] (n=jungti12@
15:09:31 Quit damaki_ (Read error: 110 (Connection timed out))
15:09:42Mooscv: check google+ there is one wad in the wiki page iirc
15:09:57MoosHello Jungti1234
15:13:40Mikachucv: it came when you bought doom ;)
15:15:02Mooshehe, I recognize the Mikachu conformist man ;)
15:15:13Mikachui only pretend
15:16:14MoosSaint Mikachu :)
15:16:32 Part LinusN
15:17:08midkayamiconn, you wondered what this fixes - the problem was that sokoban sets its own foreground colors (black text, brown boxes, etc) - but it doesn't set a background. so if one has a black background selected with white text, they'll have black-on-black text, and dark brown on black.
15:17:52Jungti1234hi Moos
15:19:01midkaymoosie :)
15:19:18cvMikachu: Oh...okay...and (i don't have doom) there are free wads, no?
15:19:33peturcv: check the wiki
15:19:34Mooshehe, hi midkay
15:19:46cvpetur: okay, thanks :)
15:21:02JdGordonPaulJam: no, the damn thing doesnt work again :'( keeps loading the same file....
15:21:21cvthanks everybody...bye
15:21:33PaulJamJdGordon: for me it changes between 3 directorys
15:21:36Jungti1234:'( my h300 is looks bad..
15:21:37Moosciao cv
15:21:55 Quit cv ()
15:22:04 Quit akaidiot (Read error: 110 (Connection timed out))
15:22:24JdGordonalthough... it seems to work from the wps tho...
15:22:26 Join _FireFly_ [0] (
15:24:04PaulJamJdGordon: sometimes it seems to choose a randon directory, but then it goes back to /-=Daten=-/Zeug/*
15:24:54*JdGordon has an idea!
15:26:54 Quit MagusG (Connection timed out)
15:28:38Jungti1234bye T^T
15:28:47 Quit Jungti1234 ("Http://www.ZeroIRC.NET Zero IRC Ver 2.9b")
15:29:13shrizzaI've occasionally been getting this black line appearing across the screen and slowly fading out.
15:29:30 Join MagusG [0] (
15:29:31shrizzaUsually when I shut off or start up rockbox.
15:29:53shrizzaA horizontal line.
15:30:18PaulJamJdGordon: If i play something from my music folder (e.g. /Musik/Damageplan/New Found Power/*) it goes first to /Musik/Alter Bridge/Find The Real/* and then it goes to /-=Daten=-/Zeug/*
15:33:19 Join TCK [0] (
15:34:09 Quit B4gder ("time to say moo")
15:34:27***Saving seen data "./dancer.seen"
15:39:53 Join ender [0] (i=ychat@
15:41:22 Quit _FireFly_ (Remote closed the connection)
15:42:27*JdGordon wants to kill rockbox's "wonderful" random gen functions :'(
15:43:13Pladaskwhat do you mean by that?
15:43:25JdGordonits not being very random at all...
15:43:39*dpro just did ugliest things to make usbvision driver compile on finally tv ;)
15:43:39Pladaskwhat is it, then? =)
15:43:46JdGordonsrand(current_tick); isnt even helping it
15:43:56dpropseudorandom ? >;->
15:44:10JdGordonno, not bloody random at all.. keep getting the same number..
15:44:36peturmaybe it's your lucky number :P
15:44:46dprooh ouch ... sorry for the >;-> then ...
15:45:07*dpro goes checking (s)rand out ....
15:45:27 Join _FireFly_ [0] (
15:45:48 Join Farpenoodle [0] (
15:47:26JdGordonunless rockbox doesnt like % ?
15:47:40JdGordonbuy rockbox i mean the coldfire cu
15:48:03dproJdGordon: does srand(clock()); help ?
15:48:15JdGordondont know if rockbox has clock()
15:48:31dproit should
15:48:59JdGordonno it shouldnt.. not on targets without the rtc...
15:49:06*dpro looks at firmware/common/random.c
15:49:22dproI just read it somewhere grepping through the sources
15:49:49PaulJamJdGordon: but why did your previous patch work? did you get the random number somehow else?
15:50:02JdGordonit didnt did it?
15:51:10amiconnJdGordon: rockbox has a really good prng
15:51:14PaulJamthe second patch with the altered line 26 worked for me.
15:51:22dproJdGordon: so you get the same number no matter if you a) seed at all b) seed with currenttick c) seed with different ints ?
15:51:26amiconnIf you get the same number over and over, you're doing something wrong
15:51:57PaulJami meant line 36
15:52:14Nibbleroi, since when can rockbox do those voice-promts?
15:52:28amiconnmidkay: Thing is that this fix shouldn't be necessary at all, because the plugin loader sets black text on lightblue background before starting the plugin
15:52:38Nibblermight collide with apples new patent? :|
15:54:25JdGordonPaulJam: ah, i figured out what changed... :'( i tinhk i can fix it :D
15:54:31 Join bluebrother [0] (
15:55:29 Quit ender` (Read error: 113 (No route to host))
15:56:40 Quit KN|stiff (Read error: 110 (Connection timed out))
15:59:57JdGordonhahah.. this makes no fucking sense...
16:00:16peturNibbler: rockbox has voice already much longer than apple :D
16:00:16*JdGordon gives up
16:00:43XavierGrJdGordon: Whatcha doing?
16:00:51Nibblerhehe, stoopid patents, once again ;)
16:00:55JdGordontrying to get a change to random dir working
16:01:41XavierGrwhy? I mean do you want something new?
16:01:52 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
16:04:10Moosrandom dir ?
16:04:48JdGordonsorry.. change to random dir at the end of the current dir... instead of just going ontot he next one
16:05:04Moosah ok
16:05:34Mooshardeep is maybe your man then :P
16:06:34 Quit MagusG (Connection timed out)
16:06:35Moosthe Rockbox playlist conceptor+the man worked on auto dir change thing...
16:06:53JdGordondid he get anywhere?
16:07:03Mooshardeep: around?
16:08:37 Join flynux_ [0] (
16:09:06MoosJdGordon: you will probably need have his agreement, or at least his opinion (on the implementation I mean)
16:09:33 Join MagusG [0] (
16:09:53JdGordonwell iv given up... so some1 else can try getting ym patch working
16:10:19Moosbut isn't this thing for after the 3.0 ?
16:10:36Moosyou have the time then :)
16:11:23JdGordon9am start tomorow.. gnite all
16:11:28 Quit JdGordon ("using sirc version 2.211+KSIRC/1.3.11")
16:13:45hardeepsorry, back
16:13:48 Join damaki_ [0] (
16:13:53hardeepMoos: you had a question?
16:14:50Mooshi hardeep, no question, just that JdGordon was talking about his current works
16:15:09Moosit seems random auto dir change
16:16:17hardeepah, yeah, i saw that he released a patch but haven't looked at it
16:16:48humulus_does the iaudio x5 remote work with rockbox?
16:16:55 Quit flynux (Read error: 110 (Connection timed out))
16:17:08Mooshardeep: he is improuving things again, let's hope he will have something good for you :)
16:17:11 Nick humulus_ is now known as humulus (
16:17:25Mooshumulus: not yet
16:17:33humulusMoos: ok
16:17:53humulusMoos: well, i should get it firt, and i no clue where to order :)
16:17:58 Join Todd [0] (
16:18:19humulusany sugesstions ...
16:18:28Mooshumulus: which model do you want, and where are you localized?
16:18:38ToddHelp! I loaded a theme on my rockbox (ipod video) and now I can't see any text.. and therefore unable to reset the font. Any way I can boot up to the default theme?
16:19:12Moosreset setting maybe
16:19:18humulusMoos: vienna, i have an iaudio X5l and i want a remote for that and a leather case would be awesome too
16:19:26 Join cv [0] (
16:19:36MikachuTodd: shut down, press menu to turn on and turn on the hold switch pretty quickly
16:20:12Mooshumulus: there are plenty on ebay... that doesn't seems difficult to find
16:20:20cvhi...again a question fo the tagcache: when i add new files to my player and force the tagcache to update, non-existing files arent there any possibility to do that?
16:20:51humulusMoos: ok
16:21:11Mooscv: not in the current version I think, not yet, but that will happen in the futur
16:21:51Mooscv: currently, you have to rebuild it for have it update (w/o the files you deleted)
16:22:42Mooshumulus: iaudio things are very expensive here in europe (Paris here), I bought all in US :)
16:22:56ToddMikachu: Thanks, you are a life saver
16:23:04ToddOr atleast a 'severe boredom' saver ;)
16:24:38 Quit Todd ("Leaving")
16:25:13 Join mareviq [0] (
16:28:03 Quit akaidiota (Read error: 110 (Connection timed out))
16:31:40 Join Kohlrabi [0] (
16:32:13 Quit damaki__ (Read error: 110 (Connection timed out))
16:34:05 Quit cv (Read error: 104 (Connection reset by peer))
16:34:46 Join ashridah [0] (
16:37:26 Join SereR0KR [0] (
16:45:14 Quit dpro ("Leaving")
16:45:56 Quit ashridah ("Leaving")
16:52:02 Join dac_adc [0] (
16:54:06dac_adchi all. Some time ago I have seen an HowTo on triple boot for the iPod (Apple fw, ipodlinux and rockbox). Still I can't find it again... Any hints?
16:54:47Mikachuput a linux.bin in / on the fat partition and hold play to boot linux
16:54:55Mikachuand of course have linux installed on a partition etc
16:55:42 Join warthawg [0] (
16:55:53warthawgis there a guide to unbricking a nano?
16:56:44Mikachudepends on what you did
16:56:46dac_adcMikachu: I already have ipodlinux on my ipod. What I want is a triple boot (I don't have rockbox yet). What I miss is the bootloader part.
16:56:53Mikachuif it's bricked, it can't be unbricked, that's the definition of bricked
16:57:01Mikachudac_adc: right, then you know everything you need to know now
16:57:39Mikachudac_adc: assuming you have a bootloader partition, a big fat partition, and a small linux partition
16:59:19dac_adcMikachu: I have all that. How can I build bootloader then? I know how to build it for Apple fw + ipodlinux and Apple fw + rockbox but I don't know how to build it for the 3 OS
16:59:42Mikachuit's the same as for apple + rockbox
17:00:08 Quit perl|wtf (Read error: 110 (Connection timed out))
17:01:20warthawgMikachu: i think it's a dead parrot now
17:01:40Mikachuwarthawg: what, specifically, does it do, and did you do to it?
17:01:47 Quit bagawk (Read error: 110 (Connection timed out))
17:02:45warthawgMikachu: i installed the 0508 rockbox (unzipped it to the partition), then unplugged it from the computer without ejecting it
17:03:07warthawgnow there is no sign of life, even if i connect it again
17:03:19Mikachutried holding menu+select for 10 seconds?
17:03:55warthawgoh, that
17:04:00warthawgnow it works
17:04:04warthawgi am sorry for the trouble
17:04:08warthawgbut thanks
17:04:32warthawgi have been holding play and select, not menu
17:04:37warthawgwhat a dummy
17:04:46Mikachuplay and select is for emergency disk mode, after you reboot with menu and select
17:07:54 Nick daurnimator is now known as daurn|afk (
17:11:27 Quit PaulJam (Read error: 113 (No route to host))
17:14:46 Join guruz_ [0] (
17:15:12 Join bluey- [0] (
17:16:02 Quit Zagor ("Client exiting")
17:18:53 Quit hardeep ("BitchX: don't leave home without it!")
17:21:50 Join akaidiot [0] (
17:25:09 Join scott666 [0] (
17:25:26 Quit scott666 (Read error: 104 (Connection reset by peer))
17:25:46 Join scott666 [0] (
17:26:50 Quit guruz (Read error: 110 (Connection timed out))
17:34:30***Saving seen data "./dancer.seen"
17:37:07 Part warthawg ("You never miss your warthawg til the well runs dry")
17:39:04 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
17:39:23 Join Febs [0] (
17:39:46 Join damaki__ [0] (
17:42:56preglowFebs: got a little sweep generator here now
17:43:38markunpreglow: do you have a link for me?
17:43:45preglownot yet, no
17:43:58markunnot presentable yet? :)
17:44:49preglowit doesn't show anything, and never yields
17:44:58preglowi might not be able to fix that last anyway
17:45:04preglowwithout spending more time on it than i care to
17:46:12markunIt does a linear sweep?
17:46:18preglowright now
17:46:28 Join Kohlrabi [0] (
17:47:44 Join gursikh [0] (n=guriskh1@
17:47:59preglowi could just use a greater block size, i guess
17:48:40 Quit petur ("*poof*")
17:49:20 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
17:50:44 Join Kohlrabi [0] (
17:51:04lostlogicpreglow: buffered or unbuffered?
17:51:46 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
17:52:15 Join Hansmaulwurf [0] (
17:52:39preglowlostlogic: unbuffered, i couldn't be arsed to import and learn all the pcmbuf stuff
17:52:48preglowthat is, it is double-buffered
17:52:52preglowbut no pcmbuf stuff
17:53:14 Join damaki [0] (
17:53:26markunpreglow: do you think a pink noise generator would be useful for adjusting the EQ?
17:53:33 Join jochen_ [0] (
17:54:10preglowin what way? you'd pretty much need a spectrum analyser to make sense of the output
17:54:18preglowand white noise would be better
17:55:10markunjlo says he can do it by hearing with pink noise
17:55:16jochen_hi, i've got a question concerning the midi playback (h1x0): would it be possible to prevent the player from powering down due to the power-down-timer during playback?
17:55:33 Quit damaki_ (Read error: 110 (Connection timed out))
17:55:47jochen_yeah, i know, feature freeze, but i think, it can be considered a bug. ;)
17:56:00preglowmarkun: yeah, but not every person can do that, heh
17:56:14preglowi sure as hell haven't got a chance
17:56:45Mikachujochen_: look in the test_fps plugin for code that does that, iirc
17:57:27jochen_Mikachu: i should have known, that i would be asked to code it myself. ;)
17:57:41jochen_Mikachu: thanks for the hint anyway
17:57:44Mikachui'm not official in any way
17:59:22markunpreglow: I can imagine why pink noise is more useful for the human ear than white, but still..
18:00:49jochen_Mikachu: any idea what i am supposed to look at in detail?
18:01:03 Join Kohlrabi [0] (
18:01:24Mikachujochen_: i think the function call is something like 'update_shutdown_timer'
18:01:32Mikachui guess i can look
18:01:53Mikachumaybe it's not there anymore
18:01:55*jochen_ cvs ups
18:02:01 Join mikearthur [0] (
18:03:19jochen_Mikachu: reset_poweroff_timer()
18:03:21Mikachujochen_: it's
18:03:23Mikachuthat one
18:04:16jochen_Mikachu: but that isn't supposed to be called by the codecs themselves, is it?
18:04:36Mikachumidi is just a plugin
18:04:58jochen_oh, sure, forget it. ;)
18:05:04Mikachuso the playback system isn't involved, isuppose that is what usually resets the timer
18:05:28jochen_yeah, i'm confused. too much sun...
18:05:39 Join ericmoritz\0 [0] (n=ericmori@
18:07:39 Quit damaki__ (Read error: 110 (Connection timed out))
18:10:39jochen_Mikachu: well, that was an easy one. patch ready. ;)
18:12:38 Quit Acksaw ("I'm off, see ya later!")
18:14:53jochen_ok, gtg, thanks Mikachu. bye
18:15:01 Quit jochen_ ("Verlassend")
18:23:36 Join Ribs [0] (
18:27:14 Part ericmoritz\0
18:30:02 Join Farpnut [0] (
18:30:13 Join dpro [0] (
18:33:14 Join damaki_ [0] (
18:34:00 Quit TCK (Read error: 104 (Connection reset by peer))
18:34:11 Join jbauman [0] (
18:39:45 Join goa [0] (
18:40:26 Quit MagusG (Connection timed out)
18:40:54 Quit goa (Remote closed the connection)
18:41:04 Join goa [0] (
18:41:30 Quit Farpenoodle (Read error: 110 (Connection timed out))
18:44:36 Join MagusG [0] (
18:46:50 Join Lear [0] (
18:47:10 Quit damaki (Read error: 110 (Connection timed out))
18:50:27 Quit goa ("Client suicide")
18:50:32 Join goa [0] (
18:50:32 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
18:50:34 Join bagawk [0] (
18:56:48 Join chang [0] (
19:02:20 Quit Farpnut ("Large sacks of cheese")
19:08:33 Quit gursikh (Success)
19:14:42 Quit chang ("So long, and thanks for all the fish !")
19:22:23 Join fraggsta [0] (
19:23:56fraggstaDoes anyone know whether the Audioscrobbler plugin *should* work on an X5. The device knows what the time is so I guess it has an RTC. Yet after patching it into a CVS checked out version of Rockbox from Sunday, the player doesn't contain the log file the plugin should write, or any menu items/other evidence that it's actually working.
19:27:48markunfraggsta: try debugging it in the simulator
19:30:19 Join PaulJam [0] (
19:32:29fraggstamarkun: I'll have to do that when I get a chance. I'm thinking of trying to join in on the X5 dev, anyway so I'd like to have a good look around the sources at some point..
19:33:06fraggstabut right now my oven is burning my pizza, oshi!
19:33:49 Quit MagusG (Connection timed out)
19:34:12 Quit bluebrother ("Leaving")
19:34:34***Saving seen data "./dancer.seen"
19:36:46 Join hd [0] (
19:37:35 Join MagusG [0] (
19:39:38 Quit mareviq ("nie wszystko dioda co sie swieci...")
19:41:03 Quit _FireFly_ ("Client exiting")
19:42:31 Join pixelma [0] (n=pixelma@
19:43:05*amiconn had an idea how to save bus bandwidth on coldfire for both playback & recording :)
19:43:48amiconnBu using the DMA autoalign feature, with reading (playback) or writing (recording) whole lines from/to memory
19:44:30amiconnI'm just testing this with s/pdif recording
19:45:23 Join theli_ua [0] (
19:45:30 Quit bluey- ("Leaving")
19:48:02amiconnIt works...
19:48:23amiconnThe question is how much we really save, and how to measure it
19:55:47preglowwhy does autoalign help at all?
19:59:14amiconnWhen SSIZE and DSIZE are different and AA is enabled, the DMA controller aligns the accesses with larger size and tries to access the larger size at once whenever possible
19:59:24amiconn-> line bursts in our case
20:00:37 Quit hd ("Client suicide")
20:08:55 Join Dgently [0] (
20:11:05preglowahh, right
20:11:11preglowthat won't be easy to measure
20:12:10amiconnIt's working fine, also for playback
20:12:41preglowyou done any comparisons on the recorded files, though?
20:12:46amiconnThe only necessary change is setting a few more bits.
20:12:49preglowit is highly improbably you wouldn
20:12:51amiconnNo, how?
20:12:54 Quit fraggsta ("Ex-Chat")
20:12:59preglowit is highly improbably you wouldn't hear the errors, but anyway...
20:13:21preglowjust like recording the same thing twice via spdif and comparing the files
20:13:35 Quit webguest73 ("CGI:IRC (EOF)")
20:13:42amiconnIt's very unlikely to start the recording twice at exactly the same sample...
20:14:23preglowyeah, you'd need to trim the resulting files so they'd fit
20:14:30preglowanywho, like i said, you would probably hear any errors
20:14:54preglowbut ok, i'll merge our patches now
20:14:56*amiconn has no idea how to align 2 such recordings
20:15:09amiconnThe offset may be thousands of samples
20:15:26preglowby using a sample editor
20:15:28preglowbut anyway
20:15:39preglowall your changes are in the one area of the patch?
20:15:44preglowi've already fixed the typo here
20:18:07scott666Is there a program that makes it easier to make a playlist for a recursive set of folders than just adding the folder in winamp and saving a playlist?
20:18:28Mikachudir /r > file.m3u or something
20:19:02scott666*that can do batch jobs
20:19:33scott666a whole bunch of folders
20:19:56amiconnpreglow: Yes, only in pcm_record.c: pcmrec_dma_start() and DMA1()
20:20:08scott666and fiddling with the command prompt is really 'easier' than just doing it with winamp
20:20:30amiconnNote that I don't think the code is ready for commit; some ifdefing for coldfire targets without s/pdif are missing,
20:20:55amiconnand somehow the recording source setting doesn't remember 'digital'
20:21:47amiconnpreglow: You can try the auto-align stuff yourself: In pcm_record.c change the 2 lines which setup DCR1 to:
20:21:54amiconnDCR1 = DMA_INT | DMA_EEXT | DMA_CS | DMA_AA | DMA_DINC | (3 << 17) | DMA_START;
20:22:09amiconnand in pcm_playback.c, change line 98 to:
20:22:10amiconnDCR0 = DMA_INT | DMA_EEXT | DMA_CS | DMA_AA | DMA_SINC | (3 << 20) | DMA_START;
20:23:13 Join blind [0] (
20:23:30amiconnconfig-h100.h should also #define HAVE_SPDIF_IN
20:23:31blindanyone who remembered my issue with my ipod: i sent it back to apple and they're sending me a new one :)
20:24:28amiconnOh, and the recording code should switch on spdif power when recording from spdif, and switch it back to the global value when exiting
20:25:21 Join damaki__ [0] (
20:26:37 Join Rondom [0] (
20:26:42 Join obo [0] (
20:32:25*amiconn loves the Firefox SessionSaver extension and its snap-back tabs feature :)
20:33:53 Quit theli_ua ("")
20:36:25crwlI like Tab Mix Plus more, I had some problems with SessionSaver
20:37:06crwlthey do basically the same thing
20:38:27 Quit Dgently ("CGI:IRC")
20:39:23 Join Thus0 [0] (n=Thus0@
20:39:29 Join gursikh [0] (
20:39:30 Quit sharpe (Read error: 110 (Connection timed out))
20:39:35 Join bluey- [0] (
20:39:55preglowamiconn: snap back tabs?=
20:40:08preglowsessionsaver i use, and it's great
20:40:25amiconnYes, that's what I mentioned...
20:41:08amiconnI already closed the tab containing your patch, and I wanted to reopen it for comparison. No need to dig for the url...
20:41:15 Quit damaki_ (Read error: 110 (Connection timed out))
20:41:37gursikhtab mix plus is the best
20:42:15PaulJami prefer tabbrowser extensions
20:43:14amiconnPaulJam: Tabbrowser extension doesn't do what sessionsaver / tab mix plus / session manager do
20:43:33gursikhas long as it's not session manager. It has such an ugly memory usage scheme, it's nasty. Wierd memory leaks all over the place (known issues)
20:43:37blindi agree yo
20:46:30 Nick blind is now known as Terminal (
20:46:52 Nick Terminal is now known as blind (
20:47:33markunamiconn: wow, it really is a nice extension!
20:47:59 Join damaki_ [0] (
20:48:44gursikhedit: I meant SessionSaver above. Have not looked into Session Manager
20:49:01scott666sessionsaver is definitely awesome
20:49:10gursikhI hate it.
20:49:13blindonly extension I use is web dev toolbar :)
20:49:55*amiconn uses a total of 7 extensions (including the one installed by the setup)
20:50:03scott666i couldnt live without adblock + filterset G
20:50:12 Join xmixahlx [0] (n=xmixahlx@
20:50:19blindadblock? filterset?
20:50:25blindjust edit your hosts file :P
20:50:42gursikhadblock is much for intelligent with filterset g
20:51:05scott666and you dont have to fiddle with it every week
20:51:08gursikhthan usuing a hosts file (which i did before usuing adblock
20:51:30scott666forecastfox is convienent too
20:51:34amiconnDOM Inspector, Adblock, Fetch Text URL, SessionSaver .2, Tabbrowser Preferences, DownThemAll!, Mouse Gestures
20:52:22scott666fasterfox is worth having too
20:53:13gursikhI have a number of extentions: stumbleupon,fasterfox,webdeveloper,disable targets for downloads, mouse gestures, ie tab, foxytunes, searchpluginhacks, customizegoogle, download status bar, google preview, menu editor, gmail manager, adblock, filterset G, flashblock, Firefox extension bakcup extrension, Tab mix plus, Performancing, Foxytunes Skin - CleanJr
20:53:20gursikhphew, too much to type...
20:55:02XavierGryay I just bought 2 optical cables for 10 euros!!
20:55:16XavierGrNow I need an optical device (except my H100) to test optical recording!
20:55:52*amiconn didn't manage to visit the electronics shop today
20:56:14gursikhgood soundcard
20:56:25preglowbut ok
20:56:25XavierGrI managed to steal a little time from my work
20:56:30preglowi think i'll do an updated patch now
20:56:52 Join damaki [0] (
20:57:01MikachuXavierGr: just point it at your lamp
20:57:02preglowahh, i need to fix the spdif power issue first
20:57:25amiconnAt least I know what components to buy tomorrow
20:57:58preglowamiconn: i'll just make an audio_get_spdif_sample_rate(void) call, then you can implement it for archos so we can have recording screen spdif sample rate info, ok?
20:58:10*amiconn will complement his diy opto->coax converter with a coax->opto converter
20:58:18 Join MarcoPolo [0] (n=MarcoPol@
20:58:41amiconnpreglow: Ok. Just don't wire it for archos, I have to find out whether it is possible to do
20:58:56blindcoax > opto lol.
20:58:58amiconn(need to check the channel status bits with different input signals)
20:59:16blindThat's like.. I'll make a dialup->cable converter
20:59:17amiconnblind: ??
20:59:36amiconnStrange comparison.... both are s/pdif
21:00:00blindim horrible with analogies.
21:00:56XavierGrwow I can see a red light coming from the cable!! :D
21:01:00amiconnpreglow: Do you know whether it's possible to monitor analog in via spdif out? (iiuc it should)
21:01:47gursikhXavierGr: Just stare at the pretty red light (right in the cornea is recomneded) as long as you can for optimum sound clarity
21:02:18Mikachuwhat's the advantage of an optical digital signal over an electrical one?
21:02:30XavierGrIs it harmfull for the eye? I thought optical light was just a LED
21:03:21preglowamiconn: sure
21:03:27 Quit damaki__ (Read error: 110 (Connection timed out))
21:03:35gursikhwell, maybe you should ask blind....
21:03:48*blind giggles
21:04:03XavierGrreally though, is it harmfull?
21:04:04gursikhsounds like he/she may have experience you might benifit from.......
21:04:35 Quit mirak (Remote closed the connection)
21:05:08 Quit RedBreva_ ("using sirc version 2.211+KSIRC/1.3.12")
21:05:14amiconnToslink uses just a led, so it's definitely harmless
21:05:16 Join mirak [0] (
21:05:20 Join damaki__ [0] (
21:05:24blindXavierGr: lmfao
21:05:32blind``really though guys. stop playin around``
21:05:41gursikhHahaha, sorry guys
21:06:25Mikachuleds can be really strong if you want to, it's just a name for the technology used
21:06:42Mikachuit's like saying a candle is just a fire so it's harmless
21:06:50XavierGrSo I remember correctly. Pheww I was starting at it for a bout 5 minutes! :D
21:06:57*Mikachu continues the bad analogy trend
21:07:25Mikachui have used a led laser in a laboration! it was 50mW or something though :)
21:07:52blindI used a 50mL laser before.
21:07:54amiconnYes, laser is a different thing
21:08:00gursikhActually I have read that optical cables can be harmful to the eye if used improperly, of course it may have been in the manual for a toslink where they guard against all possibilities of you hurting yourself and indemnifieing themselves of all liability
21:08:09amiconn(regarding to danger)
21:08:28blindyou know what's fun? optical mouse + laser pointer
21:09:16preglowamiconn: btw, it almost seems like both spdif in and out is wired to the same controller
21:09:20XavierGrI have an optical mouse but I can't see the light because they used a frequency that it is not visible
21:09:40blindwhat fun is that?
21:09:43Mikachuamiconn: it's still led though
21:11:41 Quit damaki_ (Read error: 110 (Connection timed out))
21:11:50preglowi'd imagine simulatenous transmit and receive is possible, though
21:11:51amiconnpreglow: Are the pcm output buffers in iram or sdram? (guess sdram)
21:12:03preglowamiconn: from the codecs, iram, after pcmbufifying, sdram
21:12:20amiconnI mean where the dma is running...
21:12:24preglowsdram, afaik
21:16:04preglowamiconn: btw, regarding putting spdif in signal through spdif out, i don't see how we can route the incoming clock to the output
21:17:41 Join new5guser [0] (
21:17:44preglowamiconn: unless the feed through options in the EBUxCONFIG register has the side effect of ignoring the clock field
21:18:06preglowwould make sense, i guess, but it's not the first thing i've seen in coldfire audio that doesn't make sense
21:18:34 Quit MagusG (Connection timed out)
21:19:11amiconnI would think just setting feed-thorugh should be enough
21:19:23preglowwe'll find out, i guess
21:19:36XavierGrok I don't have a clue about these stuff, but if SPDIF is digital,can the port be used to transfer data (as files)?
21:19:46amiconnSee footnote 12
21:19:48preglowtheoretically, sure
21:20:07preglowah, no reformatting
21:20:14 Quit damaki (Read error: 110 (Connection timed out))
21:20:26preglowwould work practically as well, i guess
21:20:28MikachuXavierGr: modems can transfer data over an analogue line
21:20:32 Join MagusG [0] (
21:20:57XavierGrah yes so theoritically we could use the analog port to do that
21:21:23Mikachui think someone extracted something from the ipod by playing it through the piezo and decoding it through a microphone
21:21:24preglowyou can use all ports to transfer information
21:21:29XavierGrbut as preglow said 'theoritically' I am don't know if it is doable
21:21:29Mikachui don't remember exactly what it was though
21:21:40preglowi can use the backlight to transfer information if i wanted
21:21:49 Join damaki_ [0] (
21:22:33XavierGrBut with spdif it should be easier because it is already digital and famous for its speed and bandwidth
21:23:02preglowfamous for its speed? it isn't exactly mind numbingly fast
21:23:19preglowyou could transport tops 96000*20*2 bits per second, i think
21:23:24preglowunless you can control the link layer as well
21:23:52preglowhalf a meg per second
21:23:53amiconnI guess s/pdif monitoring for analog in requires setting TXSOURCE to iis1RcvData, and CLOCKSEL to SCLK1 ?
21:24:11XavierGrthen I am talking bs!
21:24:19preglowamiconn: i'd imagine
21:25:28preglowbut ok, mborus just verified that a 48khz source still gets interpreted as 44.1khz by rockbox, even in the spdif debug screen
21:25:36preglowso i'll include my freq measure code
21:26:00amiconnMy source is properly detected
21:26:11preglowmine too
21:26:15amiconnI wonder what's going on there...
21:27:06preglowhardware bug, possibly
21:27:20preglowit is possible for the channel config register to just contain erronous bits to describe the sample rate
21:27:33preglowmost things, like dacs, etc, would just ignore that data anyway
21:30:21*amiconn just noticed that we can save some more energy on archos when someone just wants to listen to radio with no recording standby
21:30:38amiconnThen we can power off the MAS dsp core
21:31:11scott666my FMR got so awful reception the radio was basically useless
21:31:32amiconnWell, the newer Ondios are way better
21:31:49amiconnThey use the same philips fm chip as the iriver h1x0 and h300
21:32:27amiconnOld Ondios use the samsung chip also used in the fmr, and yes, its reception is awful
21:32:52 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
21:34:35***Saving seen data "./dancer.seen"
21:36:35 Quit damaki__ (Read error: 110 (Connection timed out))
21:37:17pixelmaamiconn: that's a good idea
21:44:19 Join akaidiota [0] (
21:50:29 Part lBoxy
21:51:39 Quit MarcoPolo (Remote closed the connection)
21:51:39 Quit new5guser ("CGI:IRC (EOF)")
21:55:20amiconnNow if I knew what the spdif chasnnel status bits mean...
21:57:39preglowthey should be described in specs here and there
21:57:43preglowyou could also ask linus
21:58:02preglowdoes the mas support double rate recording?
21:58:06preglowmp3 doesn't, so i guess not...
21:59:44preglowsomeone should make using whitespace in c mandatory
22:00:08Leardigging in termor again? :)
22:00:33preglowhaha, no, just ran across a xxxx?yyyy:zzzz in recording code
22:00:39preglowit's not exactly easier to read with no whitespace
22:01:09 Quit akaidiot (Read error: 110 (Connection timed out))
22:01:19 Join akaidiot [0] (
22:02:39amiconnpreglow: Doube rate?
22:03:06preglowamiconn: yeah, 64khz, 88.2khz and 96khz
22:03:15amiconnAH, no it doesn't
22:03:25 Join sockerteze [0] (
22:03:49amiconnThe MAS supports all standard mpeg audio sample frequencies, including lsf (mpeg 2.5)
22:03:57 Join petur [0] (n=petur@rockbox/developer/petur)
22:04:17amiconnMAS s/pdif only supports mpeg 1 sample rates
22:04:37 Quit akaidiota (Read error: 110 (Connection timed out))
22:06:07preglowamiconn: then you should be able to use the freq detection code in the debug screen
22:06:16preglowamiconn: the status of the channel config reg should be the same
22:06:19 Join qwx [0] (
22:06:55amiconnI'll just compare them
22:07:06amiconnThe mas only gives 16 bits of the channel status
22:07:21preglow16 bits is all there is, afaik
22:07:46preglowhmm, no
22:07:48preglowthe debug code uses more
22:14:57 Quit qwm (Read error: 110 (Connection timed out))
22:17:16 Join WileE79 [0] (
22:18:08WileE79Hey everyone...
22:18:25 Join B4gder [0] (
22:18:31 Part B4gder ("Run away!!!")
22:19:05WileE79is it ok to ask questions regarding jukebox recorder and rockbox???
22:19:55markunWileE79: yes, just ask
22:22:34WileE79my jbr doesn't work with batteries anymore (kills them regularly) so I was wondering if there is way to use it with no batteries inside, just via the charger plugged into it. It seems like this doesn't work with the original firmware so I was wondering if there is a way to do this with rockbox
22:22:51amiconnpreglow: Hmm, seems we're unlucky :(
22:22:56 Join sharpe [0] (
22:23:16MikachuWileE79: probably not
22:23:38amiconnThe coldfire delivers the first 32 channel status bits, in reverse order (bit 31 of EBURCVCCHANNEL is bit 0 of channel status)
22:24:13amiconnThe MAS delivers the first 16 channel status bits only, in proper order (bit 0 of register 0x56 is bit 0 of channel status)
22:24:47WileE79:-( this means I have to throw away an almost fully functional mp3 player *buhu*
22:24:55amiconnFrequency is bit bit 24..27 :(
22:25:06MikachuWileE79: maybe you could connect some capacitors, but don't quote me on that
22:25:25amiconnWileE79: YOu mean batteries don't charge inside the recorder?
22:25:38amiconnWhat if you charge them in an external charger?
22:25:56preglowamiconn: what, and they just drop them?
22:25:58WileE79No they charge fine but after 2 or 3 days they are empty even if I don't use the jbr
22:26:01preglowamiconn: btw, did you find a description of all the bits?
22:26:17amiconnI just used the spdif analyzer debug code
22:26:34preglowah, ok
22:26:43preglowsounds weird they don't give you all bits
22:27:06amiconn..and compared the 2 values I was able to get: CD in DVD player, DVD in DVD player, and no disk in DVD player
22:27:12amiconn*3 values
22:27:25scott666WileE79: whats the problem then?
22:27:26amiconnEven coldfire doesn't give all bits
22:27:49preglowbtw, currently i write the sample rate together with the header at the start of the recording. this will work out rather badly if people start the recording with no cable inserted. how should we deal with that?
22:28:03MikachuWileE79: put in a piece of plastic between the battery and the connector
22:28:46WileE79Well, I put fully charged batteries in my JBR and after 2 days they are empty...they also discharge if I remove them from the device and just have them lying around
22:29:05 Join zigford [0] (
22:29:13WileE79It looks like they get killed during charging
22:29:19preglowamiconn: btw, could you please check i merged your patch alright? -
22:29:33WileE79after the first charging circle
22:29:36amiconnWileE79: That may happen with somewhat used NIMH cells, especially if they are high-capacity and no-name (in combination)
22:30:06amiconnWileE79: Hmm...
22:30:14amiconnfirst cycle?!?
22:30:18 Join lookup [0] (
22:30:48WileE79I already fried 4 sets of batteries (tru they were high capacity) two sets from energizer
22:31:41WileE79ups...yes cycle...
22:32:19amiconnpreglow: config block version needs to be bumped one more (already is at 41)
22:32:36scorcheWileE79: how high capacity are we talking?....i have had issues with batteries above 2400
22:32:40preglowamiconn: ahh, yeah
22:32:49preglownot the finished patch anyway
22:33:05lookupi have a question about ipod on rockbox. anybody that knows what they are talking about can surely message me
22:33:14WileE79if I remember correctly 2200 - 2500
22:33:31preglowlookup: or you can just ask a question right here
22:33:33scorchelookup: just ask the question...
22:33:59lookupif i install rockbox on my ipod is there a possibility that i could fubar it?
22:34:07preglowthere's always a possibility
22:34:12preglowbut in the case of ipods, it's very, very, very small
22:34:17preglowno one has done it so far
22:34:43lookupwhen i do change the fw, i will have the oppurotunity to update my songs via Sharepod?
22:34:44preglowno matter how much you mess it up, you can always use the apple ipod restorer program
22:34:57preglowwith rockbox, you just copy files
22:35:00preglowno extra program needed
22:35:08lookupo really
22:35:08lookupthat's nice
22:35:47lookuphow long does an installation like this take?
22:36:22 Quit Lear ("Chatzilla 0.9.73 [Firefox]")
22:36:59Mikachua couple of minutes if you've done it before, probably longer otherwise
22:37:13amiconnBah, firefox and its caching :/
22:38:00gursikhdisable it
22:38:19WileE79:-) better than IE6 caching (IE7 is doing better)
22:39:00Mikachu[lookup] is this better than ipod fw or is it worse? in quality i mean.
22:39:01amiconnpreglow: tabs slipped in....
22:40:23WileE79So do you guys think its worse to give it another shot with cells with lower capacity before I go and get a new Gmini?
22:40:49 Quit xmixahlx ("blah blah blah")
22:40:56 Quit SereR0KR (Remote closed the connection)
22:41:13preglowamiconn: yeah, know, like i said, i'm only looking to see if i got your patch right at this point
22:41:25 Join qwm [0] (
22:41:27preglowit's not done
22:42:05sharpewell, i've made a graph of ipod video battery usage/statistics.
22:42:46sharpeit's a pretty curve
22:43:00amiconnpreglow: Looks like you did
22:43:40 Join hardeep [0] (i=hardeeps@otaku.freeshell.ORG)
22:44:06sharpeseems to last about 19000 seconds, or about 5.2 hours...
22:44:49 Quit dpassen1 ()
22:45:26sharpevoltage level just... drops insanely before it dies...
22:46:00 Join SereR0KR [0] (
22:46:51Mikachusharpe: do you want to show us this curve or should we reassemble it from your description?
22:47:44sharpei was actually going to try for a formula... but... okay...
22:48:57Mikachulagrange interpolation pls
22:50:44 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:51:10markunsharpe: does it look anythink like the iriver battery curves?
22:51:25 Quit amiconn (Nick collision from services.)
22:51:25 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:51:56 Quit goa ("Client suicide")
22:52:06sharpei have no idea what the iriver battery curves look like.
22:54:29 Quit bluey- ("Leaving")
22:54:29 Join goa [0] (
22:54:56 Quit qwx (Read error: 110 (Connection timed out))
22:55:16sharpeslightly sharper dropoff at the bottom, it seems.
22:56:20 Join Gargamale [0] (
22:56:36preglowamiconn: what would be the better approach? duplicating the spdif power code in pcm_recording.c, or moving the whole power management to recorder/recording.c?
22:56:40preglowthink i vote for the last
22:56:51 Join damaki__ [0] (
22:57:16preglowat least there i know if the power is supposed to be on or not
22:57:21amiconnJust leave the power code in power.c, and use it...
22:57:51preglowno code in power.c can tell me if the power is supposed to be on or not
22:57:57preglowor if the power is enabled currently
22:58:13*amiconn wants access to byte 3 :-(
22:58:37 Join dpassen1 [0] (n=dpassen1@resnet-236-163.resnet.UMBC.EDU)
22:58:43amiconnThe app layer knows whether spdif power should be on by default
22:58:56amiconnWhile recording from spdif it needs to be always on
23:00:06 Quit SereR0KR ("XChat Aqua")
23:00:29preglowyes, and that's why i want to do it in recording.c
23:00:49preglowfirmware layer does not know whether to leave it on when deselecting spdif
23:01:11amiconnNo, but app layer shouldn't switch it itself
23:02:59preglowok, so i'll need to read the status of the pin on entry of the recording screen to find out if it should stay enabled or introduce an enabled variable in power.c?
23:03:04preglownone of them sounds very elegant
23:06:40 Join dj-fu [0] (
23:06:43preglowwhat about just having pcm_record.c enable it without checking, then having recording.c call spdif_power_on() if checking the settings struct says it's needed
23:07:04preglowlinuxstb: here?
23:08:11amiconnYou don't need to read status on entry. Recording screen is app layer, and can read the global settings value
23:08:52 Join Daishi [0] (
23:08:57amiconnHowever, in order to save energy when spdif is not needed/wanted, recording.c would have to tell pcm_record.c what the default state should be
23:09:53preglowrecording.c knows what source is selected, yes? why can't it just switch itself?
23:09:59preglowah, no, that's sound_menu.c code
23:10:18*preglow checks
23:11:44preglowrecording.c does the set_recording_options() stuff itself, so it could work without needing to tell pcm_record anything
23:11:45 Quit damaki_ (Read error: 110 (Connection timed out))
23:12:13 Quit Hansmaulwurf ("( :: NoNameScript 3.81 :: )")
23:12:36 Join qwx [0] (
23:13:56 Quit nobelium- ("( :: NoNameScript 4.02 :: )")
23:16:04 Quit Rondom (Connection timed out)
23:17:37 Join ashridah [0] (
23:19:43 Part Pi
23:24:57 Quit qwm (Read error: 110 (Connection timed out))
23:25:25 Quit Ribs ("Leaving")
23:26:52 Quit jbauman (Read error: 110 (Connection timed out))
23:32:38amiconnpreglow: Looks like we cannot get the sample rate from spdif input on archos, at least not for mp3 recording
23:32:54amiconnFor pcm recording it will work...
23:33:02preglowamiconn: we could parse headers coming out of the mas
23:33:08amiconnBut we can report the sample rate in the recording screen
23:33:30amiconnIt will just not update while recording is paused
23:34:03amiconnWe don't even need to parse, we can read the header from 2 MAS status memory cells
23:34:23 Join TCK [0] (
23:34:39***Saving seen data "./dancer.seen"
23:35:14amiconnSince the MAS records for a short time (a few ticks) during recording init, the reported sample rate will most often be correct (in case the source is already present when entering the recording scren)
23:37:09 Join Doomed [0] (
23:37:51 Join hd [0] (
23:38:24 Quit Febs ("CGI:IRC (EOF)")
23:38:43preglowamiconn: but yeah, what do you say to my wav header sample rate question? i don't think you answered it
23:39:32 Quit lookup ()
23:40:24 Quit hd (Client Quit)
23:40:40 Join hannesd__ [0] (
23:40:56sharpeahah... quartic function for battery level after however many seconds...
23:43:48amiconnpreglow: Hmm, there's a similar situation on archos. We read the 'template' header shortly after starting the recording (200ms) and use that to calculate our xing header
23:44:42amiconnSo the xing header will be messed up if the sample rate changes during recoridng
23:44:53amiconn..but I think that's the user's fault
23:45:00preglowyeah, i think we'll just say that for now
23:45:07preglowrecording will need a overhaul some time in the future anyway
23:45:18 Nick hannesd__ is now known as hannesd (
23:45:18 Join Gargamal1 [0] (
23:45:21preglowi'll just do a note of it in the manual
23:45:28amiconnWith mp3 it will still play correctly (with a slight glitch), but with wav there's no way to tell, so pitch will be off.
23:45:38preglowat least it's very easy to fix
23:45:42preglowyou just alter the wav header
23:45:53amiconnNot if it changes inbetween
23:46:02preglownono, but then you've got problems either way
23:46:09preglowyou can't fix that in a wav except splitting to a new wav
23:46:20preglowand i can't be bothered to do that now
23:46:44amiconnWhen we add the recording framewalker, we should make the recording engine split files on frequency change
23:48:01 Join piroko [0] (
23:48:24pirokonot sure if you guys know, but 4g support has been broken for a while now
23:49:02pirokowon't boot period
23:50:07preglowyou sure?
23:50:12preglowi would have thought linuxstb would have noticed it
23:50:19pirokopositive. i've tested it on two different ipods now
23:50:23 Quit BHSPitLappy (Connection timed out)
23:50:44pirokoused the mac instructions. it should still work though, right?
23:51:08pirokothe bootloader works and everything
23:51:13pirokorockbox freezes at the logo
23:51:42 Quit Gargamale (Read error: 110 (Connection timed out))
23:52:00sharpethat's not good, is it? :-D
23:52:25 Join bluebrother [0] (
23:52:30XavierGrcan someone close this bug report:
23:52:35 Nick Gargamal1 is now known as Gargamale (
23:52:45XavierGrThere is already one present for the same bug
23:52:53pirokosharpe: nope :(
23:53:11 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
23:53:51markunXavierGr: where's the other one?
23:54:27XavierGrone moment please
23:55:44tucozanyone know how to check out a certain revision of a file from cvs?
23:57:50 Join damaki_ [0] (
23:58:33bluebrothertucoz: cvs co -r <tag> IIRC
23:58:37tucozbluebrother, sorry
23:59:00tucozI found out. It was cvs up -r <tag>

Previous day | Next day