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 2013-07-01

00:00:19 Join belak [0] (~belak@facebook/engineering/belak)
00:00:38 Quit z180 (Read error: Connection timed out)
00:01:18 Join z180 [0] (~chatzilla@ip-2-202-150-178.web.vodafone.de)
00:09:54 Quit belak (Quit: belak)
00:31:01 Quit n1s (Ping timeout: 246 seconds)
00:36:55 Part SrRaven
00:50:05 Quit ender` (Quit: Eighty percent of all people consider themselves to be above-average drivers.)
00:52:37 Quit Rondom (Ping timeout: 252 seconds)
00:53:12 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
00:55:03 Quit z180 (Remote host closed the connection)
01:00
01:00:49 Quit Rower (Quit: Hmmm...)
01:06:36 Quit liar (Ping timeout: 268 seconds)
01:23:28 Join _jhMikeS_ [0] (~jethead71@d192-24-174-117.try.wideopenwest.com)
01:23:28 Quit _jhMikeS_ (Changing host)
01:23:28 Join _jhMikeS_ [0] (~jethead71@rockbox/developer/jhMikeS)
01:23:28 Quit jhMikeS (Disconnected by services)
01:23:29 Nick _jhMikeS_ is now known as jhMikeS (~jethead71@rockbox/developer/jhMikeS)
01:24:39 Join scorche [0] (~scorche@rockbox/administrator/scorche)
01:24:46 Quit scorche` (Read error: Connection reset by peer)
01:48:17***Saving seen data "./dancer.seen"
02:00
02:10:42 Quit lebellium (Quit: ChatZilla 0.9.90 [Firefox 23.0/20130625125232])
02:16:22fs-bluebotBuild Server message: New build round started. Revision bb87590, 217 builds, 20 clients.
02:17:10 Quit sakax (Remote host closed the connection)
02:23:18fs-bluebotBuild Server message: Build round completed after 416 seconds.
02:23:18fs-bluebotBuild Server message: Revision bb87590 result: All green
03:00
03:04:48 Nick Scr0mple is now known as Scromple (~Simon@161.43.73.67)
03:48:21***Saving seen data "./dancer.seen"
04:00
04:03:32 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:03:32 Quit amiconn (Disconnected by services)
04:03:32 Quit pixelma (Disconnected by services)
04:03:33 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:03:35 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:03:35 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:56:29 Join Rower [0] (husvagn@v-413-alfarv-177.bitnet.nu)
05:00
05:04:29 Quit [7] (Read error: Operation timed out)
05:07:06 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:48:22***Saving seen data "./dancer.seen"
05:51:43 Join belak [0] (~belak@facebook/engineering/belak)
06:00
06:30:34 Quit belak (Quit: belak)
06:33:57[Saint]Does anyone know of a nice set of media icons with a permissive license?
06:34:31[Saint]Preferably with shuffle/repeat icons, and the usual play/pause/stop/seek icons.
06:35:21[Saint]Oh, also, ...with SVg sources?
06:35:27[Saint]*SVG even
06:36:03[Saint](may seem off-topic, but its for a cabbieV2 replacement demo)
06:42:59 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
07:00
07:18:01 Join polemon [0] (~polemon@g224150003.adsl.alicedsl.de)
07:18:35polemonhello, I have an iPod nano 2nd gen, and according to what's on the website, USB is broken
07:18:54polemonI have a rather current dev build on my ipod, and USB seems to work.
07:19:01polemonis that information outdated?
07:21:38 Quit Epicanis (Quit: why am I still awake?)
07:24:27 Quit polemon (Ping timeout: 260 seconds)
07:42:22 Join mortalis [0] (~kvirc@213.33.220.118)
07:43:58 Quit amiconn (Remote host closed the connection)
07:43:58 Quit pixelma (Read error: Connection reset by peer)
07:44:51 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
07:44:52 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
07:48:23***Saving seen data "./dancer.seen"
08:00
08:06:54 Join polemon [0] (~polemon@92.224.146.120)
08:14:22 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
08:15:25 Join LinusN [0] (~linus@giant.haxx.se)
08:20:07[Saint]polemon: "rather current dev build" is quite meaningless, I'm afraid.
08:20:17[Saint]Could you actually report the version, please?
08:20:41polemonthe one from two days ago, I don't remember the rather cryptic SVN hex number
08:21:50[Saint]You don't need to.
08:21:59polemonI mean either it just works, or I don't know what to look for
08:22:09[Saint]debug (keep out!) > Rockbox Info will report the version.
08:22:18polemonok, wait a second
08:24:24 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
08:24:50polemonAh, ok, I think I found it: 662f757-130628
08:25:06[Saint]Thank you.
08:25:21polemonSystem > Rockbox Info
08:25:25polemonthat's where I found it
08:25:43[Saint]Ah, sorry, my mistake. I thought it was in the debug menu.
08:25:47[Saint]Thank you anyway.
08:25:54polemonheh, no problem
08:26:01[Saint]Though, I am rather more afraid of the situation now that it works.
08:26:08[Saint]I would much prefer it didn't.
08:26:18[Saint]Bugs that magically fix themselves are bad news.
08:26:19polemonhmm, I can do a test or to, if you want me to
08:26:23 Quit fs-bluebot (Ping timeout: 252 seconds)
08:26:45[Saint]If it doesn't panic immediately on USB insertion, then you aren't seeing the bug.
08:26:55polemonok, I'll check again
08:27:10[Saint]If it mounts at all, and you can transfer, then you're definitely not seeing it.
08:27:23[Saint]Which is "good", but rather suspicious.
08:27:42 Join fs-bluebot [0] (~fs-bluebo@g231121074.adsl.alicedsl.de)
08:28:12 Quit bluebrother^ (Ping timeout: 276 seconds)
08:30:01polemonok, it works like this:
08:30:31polemonif the thing is running, and I insert the USB cable, sometimes it panics, sometimes it does nothing, but panics when I remove the USB cable
08:30:49polemonbut, when the device is not powered, and I insert the cable AND THEN boot the iPod, it works as expected
08:31:05polemonif I remove the cable then, it resumes to normal operation
08:31:18polemonnow lets see what happens if I reinsert the cable...
08:31:23[Saint]Thank you. This is somewhat helpful information.
08:31:34[Saint]s/somewhat/quite/
08:32:37 Join olspookishmagus [0] (~pookie@host-84-205-241-1.cpe.syzefxis.ote.gr)
08:32:55 Join ender` [0] (krneki@foo.eternallybored.org)
08:33:19polemonok, when I reinsert the cable, the iPod went into the USB-Mode, but some weird things appeared in the lower right corner of the screen
08:33:31polemonthe iPod is charging, but doesn't mount
08:33:42polemonwhen I remove the cable the iPod stays in the USB mode!
08:34:12polemonthe clock is still running, but the buttons don't work
08:34:49*[Saint] prods TheSeven - can you explain the behaviour defined above?
08:34:50polemonafter a reboot it's back to normal
08:35:30polemonok, I tested it again: when I insert the USB cable WHILE it's already running, I get the panic
08:35:36[Saint]Well, the bug seems to have slightly lessened its severity...but I can see no obvious reason as to why it should do so.
08:35:43polemonwhen the cable is inserted before it boots, it goes into the USB mode as expected
08:36:13polemonand then resumes normal operation when unplugged
08:36:34[Saint]The last time I tested this, which was many moons ago, USB didn't work under any circumstances whatsoever.
08:36:49polemonexcept booting the device in Disk mode?
08:37:01polemondisk modes seems always to work
08:37:09[Saint]Correct. That is always expected to work as it isn't under our control at all.
08:37:16polemonfirst time I uploaded the dev version, it wouldn't boot
08:37:44polemonI though I've bricked it, but then booting into disk mode and loading the firmware again, worked
08:37:56[Saint]Hmmmmm...this is curious. I cannot replicate your behavior.
08:38:16polemondoes it have anything todo with the themes, maybe?
08:38:26[Saint]Which bootloader are you using? Are you using the Rockbox bootloader, or an iLoader/emCORE version?
08:38:38polemonthe one that came with rockbox
08:39:00polemonthe utility has this option, I ticked the checkbox where it said "install bootloader"
08:39:17[Saint]That /may/ have an effect on my situation, but by the time Rockbox boots whatever loaded it _shouldn't_ be relevant.
08:39:45[Saint]I will revert back to the Rockbox bootloader this evening and check this.
08:40:14[Saint]And, yes, although it is technically "impossible" for it to do so, themes very well could have an effect on this behavior.
08:40:37[Saint]Preferably all tests should be done with the system default, cabbieV2. Just so everyone is on the same page.
08:41:38polemonok, I've just selected cabbieV2
08:41:42polemonlet's see...
08:42:41polemonok, now the iPod can't get out of the USB mode again
08:43:00polemonit still displays the USB plug, but the iPod isn't plugged in anymore
08:45:14polemonok, now, the iPod didn't mount, but after removing the cable the Rockbox got out of the USB mode..
08:45:21polemonseems to have a mind of it's own, really
08:45:24[Saint]Curious. And, the behaviour is the same ragrdless if the iPod is shutdown, or running, during USB insertion?
08:45:52polemonthis was after reboot, then inserted, then unplugged
08:45:57[Saint]It may also be helpful to know what the theme was prior to changing back to cabbieV2.
08:45:59polemonit didn't panic, but didn't mount
08:46:18polemonit was a modification of electricbarsofcolor I did myself
08:49:17polemonok, now it seems to be in a state, that it senses being plugged into the USB port, and resumes operation after unplugging but doesn't mount
08:49:53[Saint]Still with cabbieV2?
08:49:58polemonyes
08:50:31[Saint]I know this is a lot to ask, and rather tedious, but could you please check if normal operation (or, the somewhat normal operation) resumes when you go back to your custom theme?
08:50:54[Saint]Thank you very much for this, by the way.
08:51:23polemonnot a problem, it's just sticking the USB plug in and out for me
08:51:26polemonand running dmesg
08:53:30polemon[346184.085112] usb 2-5: USB disconnect, device number 4 [346198.030090] usb 2-5: new high-speed USB device number 5 using ehci-pci [346203.139223] usb 2-5: device descriptor read/all, error -110 [346203.240961] usb 2-5: new high-speed USB device number 6 using ehci-pci
08:53:48polemonthis is particularily interesting: [346203.139223] usb 2-5: device descriptor read/all, error -110
08:57:24polemon[Saint]: you still there?
08:57:36[Saint]I am, yes.
08:58:08[Saint]I _think_ that is an ehci bug that seems to be present in a lot of distributions.
08:58:15[Saint]Easy to check for.
08:58:43polemonhttp://pastie.org/8098987 <−−- I've left the iPod connected to the computer for some time, and this is what dmesg says
08:58:58[Saint]Are you aware of how to disable USB 2 (temporarily?
08:59:18polemonuh, I probably have to rmmod a module?
09:00
09:00:21[Saint]modprobe -r ehci-hcd should do the trick. Though I'm not really sure how useful this is, as it isn't really much of a solution at all.
09:00:47[Saint]It would be much more interesting to know *how* we trigger this bug, but that is beyond me.
09:01:13polemonmodprobe: FATAL: Module ehci_hcd is builtin.
09:01:41[Saint]Gah. Nevermind.
09:02:21polemonhmm, it used to work, though
09:02:53polemon[346731.517935] usb 2-5: reset high-speed USB device number 30 using ehci-pci
09:03:03polemonnow I don't get the errors, but the device doesn't mount
09:03:10[Saint]The only other time I have been able to reproduce this behavior is with some Kingston USb thumb drives formatted as FAT32
09:03:32[Saint]I have no idea what triggers it.
09:03:34polemonnow I've disconnected and it returned to normal operation
09:04:16polemonit seems to be quite dificult to track down, as sometimes it mounts, sometimes it doesn't sometimes it triggers errors, sometimes it doesn't do a thing
09:04:32polemonlooks to me like either a timing problem or a protocol error
09:05:06[Saint]My guess is bad memory addresses.
09:05:22[Saint]But, its just a guess, and probably the result of a subtle bias.
09:05:28polemonok, but why does it work sometimes, and sometimes not?
09:05:46polemonand sometimes it triggers the PANIC and sometimes it doesn't as it does not
09:05:50polemons/not/now
09:06:57[Saint]Its only a guess, but I think it may have something to do with loading/unloading the USb screen, which may not necessarily be in the same address each time, if I understand correctly.
09:07:13[Saint]But the buffer system in place is insanely complex, and I get lost in it easily.
09:08:05[Saint]I could, also, be completely wrong :)
09:08:17TheSevenpolemon: if it panics, what is the exact panic message?
09:08:48polemonerm, wait a sec, got to get it panicing again...
09:08:53[Saint]AH, yes, that's something I should've asked. I only ever used to get garbage addresses in the panic.
09:09:39polemonerm, it won't panic
09:09:49[Saint]Thanks for stopping by, TheSeven. This curious, seemingly new behavior intrigued me.
09:10:09[Saint]And its no secret this is far too low level for me :)
09:11:12polemonI can't get it to panic...
09:11:12TheSeventhemes could very well have an effect, because the ipod nano USB problem looks like the whole handshaking between the USB and main threads is involved somehow... something seems to go wrong while switching to USB mode, which probably happens before something is loaded if the cable is plugged during boot
09:11:27TheSevenso it could even be the database and amount of music files etc. that could have an effect on this
09:11:44[Saint]Hmmm. Good point.
09:11:45polemonso it is a timing/protocol error?
09:12:12polemonok, third time I've tried: can't get it to panic
09:13:10[Saint]Heh. Typical. Something fails repeatably, except when you *want* it to.
09:13:33TheSevendevice descriptor read/all, error -110 means that it somehow survived the first descriptor transfers and stopped responding at some point
09:14:05[Saint]that's either a coincidence, or a reasonably common *nix bug.
09:14:25TheSeventhe whole chatter after that is just the typical "this device doesn't respond AT ALL" thing, with the ipod being locked up somewhere
09:15:16polemonindeed, if the -110 error appears, the iPod becomes unresponsive
09:15:29[Saint]I can get my desktop to behave the same way with some kingston USB thumb drives. Killing ehci "fixes" it.
09:15:40[Saint](on Ubuntu and Fedora)
09:16:17[Saint]It only happens when its formatted as FAT32 too...but I am unsure if this is a cause or a coincidence.
09:16:30TheSevenwell, what kind of panic was it?
09:16:38polemonI'm using fedora as well
09:17:07polemonTheSeven: no panic, the screen just get's stuck, and the buttons don't work, but the ipod doesn't display the PANIC text or anything
09:17:19polemonI can still see the last screen of the Rockbox menu
09:17:54TheSevenwell, I mean the panic that you got at some point
09:18:07polemonI honestly can't get it to panic
09:18:19polemonit either just locks up, or only does not mount
09:18:22polemonbut no panic
09:18:26TheSevenbut you did earlier? do you remember what the panic message looked like, roughly?
09:18:44polemonI'm sorry, no
09:18:53polemonbut I'll try to trigger the panic message again
09:18:55TheSeven[Saint]: are you aware what the typical kind of USB insertion panic is?
09:19:21TheSevenbecause on my device I usually just get lockups as well, which makes this thing hard to nail down
09:19:57[Saint]I used to get stkov nand panics with a garbage address.
09:20:03polemonoooh, another error:
09:20:09[Saint]I can't get this thing to panic either, dammit.
09:20:30polemonwhile booting up: ATA error 1
09:20:31TheSevenhm, we really need to be able to run rockbox virtualized on these devices, in user mode instead of supervisor mode, and with a little protected GDB stub accessible over serial or something
09:20:33[Saint]and yes, I'm aware that stkov nand panic makes absolutely zero sense :)
09:20:42[Saint](in this scenario)
09:21:47polemonno panic again
09:21:55polemonbut I see a lot of these after inserting:
09:21:57TheSevenhm, a stkov nand has typically exactly one cause: the nand is about to power down and calls the ata idle callback to flush some config/state files to the flash, and something in there uses too much stack
09:22:00polemon[347907.888651] usb 2-5: reset high-speed USB device number 49 using ehci-pci
09:22:04polemonit constantly resets
09:22:52TheSevenso that might very well be related to rockbox entering USB mode... 3 things to try: 1. increase NAND stack size, 2. completely disable NAND powerdown, 3. disable NAND powerdown while in USB mode
09:22:54polemonno panic, ever
09:23:22polemonI'll enable my theme that I've used before, see if that makes any difference
09:23:27[Saint]TheSeven: thanks for the clues. I shall attempt to have a look this evening.
09:24:22TheSeventhose reset messages are good news, those mean that the ipod is still alive and we are "just" seeing the kind of behavior that causes things to fail with USB2 sometimes (the root cause seems to be some race condition in the USB driver which is triggered by concurrent EPO and BULK endpoint access)
09:24:58 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:25:48TheSeventhis is the typical "mounts after half an hour" behavior, which we're also observing with UMSboot recently
09:26:06[Saint]Oh. Sorry, slight memory failure. It was stkov usb that gave the garbage addresses.
09:26:16[Saint]stkov nand doesn't print an address iirc.
09:26:36TheSevenI really need to start another try at nailing the UMSboot version of that bug down, but I won't have time for that before the end of this month
09:26:38[Saint]But, stkov nand was the most common panic I saw.
09:26:54TheSevenstkovs should all behave the same?
09:27:30[Saint]I don't recall stkov nand printing an address at all...but, this may be me misremembering.
09:27:43[Saint]Things would be a lot easier if I could get this thing to panic now.
09:27:52[Saint]But it seems to be completely ignoring USb.
09:28:07TheSevenanother thing possibly worth trying: connect USB during boot and check how it behaves for a couple of times, with varying theme/core configurations
09:28:24TheSevenwith a really simple, and an insanely complex theme, both in terms of memory usage and drawing effort
09:28:35polemonI'm sorrt I can't get it to panic
09:28:38TheSevenmight also be worth trying to enable/disable the database and dircache
09:28:51polemonit either does nothing, or locks up
09:29:08TheSevenpolemon: that's the typical mess that I'm used to
09:29:18TheSeven(with cabbiev2, didn't try other themes)
09:29:51[Saint]Mine is ignoring USB *completely* now...
09:29:58[Saint]Not even charging. Huzzah!
09:30:33[Saint]Yay for things that are practically impossible to track down and don't behave how you want when you want them to!
09:30:41TheSevenhm, I think we should really start to instrument this somehow, and preferably in a way that is able to dump data across a reboot
09:30:58TheSevenor try to split things down an start with a simpler environment
09:31:16TheSevenmaking sure that the USB stack itself works before doing any of the USB mode entry/exit handshaking
09:31:40TheSevenmight possibly be worth doing a bootloader USB build for nano2g, and checking how that behaves
09:31:50TheSevenshould act much more predictably
09:32:32[Saint]Ooooh. I never thought of that. That's a good idea.
09:33:39polemonhow many people are currently working on Rockbox?
09:34:15[Saint]That's rather hard to say. Actively? A few dozen?
09:34:23[Saint]In total, several hundred.
09:34:26polemonoh that many
09:34:47polemonI was expecting something along of maybe five to ten people actively
09:35:28[Saint]http://git.rockbox.org/?p=rockbox.git;a=blob_plain;f=docs/CREDITS;hb=HEAD
09:35:46[Saint]that's the entire project.
09:36:07[Saint]Active committers number a couple of dozen, at a guess.
09:36:39polemondo you guys plan to add more devices to the list?
09:37:01[Saint]There are no plans, per se.
09:37:12[Saint]If someone wants to work on a new port, they will.
09:37:55[Saint]There should be a few devices from Creative joining the lineup reasonably soon.
09:38:30polemon[347907.888651] usb 2-5: reset high-speed USB device number 49 using ehci-pci <−−- I have this mp3 player as well
09:38:41polemoninterestingly, it supports Ogg/Vorbis out of the box
09:38:58[Saint]What device is this?
09:39:12polemoniBeat Organix
09:40:09[Saint]I can never stop giggling about the iBeat Blaxx
09:40:18[Saint]...poor, poor, poor naming choice.
09:40:28polemonomg, I believe it was quite close to ending the company
09:40:30[Saint]Anyhoo.
09:40:39polemonTrekStor was in pretty bad shape at that poing
09:40:44polemonpoint*
09:47:33[Saint]A port for the Organix is certainly possible.
09:47:47[Saint]It "just" needs someone to do the work.
09:48:24***Saving seen data "./dancer.seen"
09:49:42polemonno idea how to port that
09:50:10polemonI'll try and add codec2 decoding to Rockbox, probably, but making a port, I wouldn't know where to start
09:51:26[Saint]That _might_ be interesting for audiobooks et al.
09:51:49polemonwell, I have another reason actually
09:52:17polemon1. I like the idea of Codec2, and I wouldn't want to see it only used on embedded devices for radio communications
09:52:35freqmodi have created support for codec2 for rockbox
09:52:36[Saint]But I rather suspect that Opus will kill a lot of the need for many codecs.
09:52:41polemon2. I'd like to see how well the speech codec performs on relatively low-quality devices like that
09:52:48freqmodit is on my home computer, however it does only work on the simulator
09:52:56freqmodlast i checked codec2 was only floating point
09:53:14[Saint]Yes, I believe you're right.
09:53:14polemonwell, codec2 is still a long way from being finished
09:53:23polemonalso development kinda seemed to have stalled a bit
09:53:32freqmodi will make a patch if/when somebody converts it to fixed point
09:53:42[Saint]Possibly in no small part due to Opus removing the need for it.
09:53:52polemonfor codec2?
09:53:56[Saint]Indeed.
09:54:02polemonerm, no, not really
09:54:07[Saint]How so?
09:54:10freqmodit is supposed to support 800bps voice if optimized for size
09:54:11polemonthere is a rather good presentation by David Rowe
09:54:14freqmodopus supports 8kbps
09:54:37freqmodnow it is optimized for rf roubustness at 1.6kbps
09:54:41polemonit works quite good at 1400bps by now
09:54:49[Saint]Whoah...holy, codec2 goes that low? I must've misread in the past.
09:55:13polemonthe c2enc lets you encode down to 1200bps, but it doesn't work too well
09:55:36polemonbut for digital SSB communication, it is pretty much the way to go
09:55:58polemonat 1200bps, you can't recognise someone's voice anymore, but you still understand the spoken words
09:56:34polemonI am from a CB-radio background, so it kinda interests me in that way
09:57:48[Saint]Hmmmm. Yes. I must've misread. I didn't think codec2 had such a low bitrate.
09:58:22[Saint]I would still like to think that Opus will kill the need for so many specific codecs.
09:58:42polemonwell, if it kills AAC, I'd be quite pleased
09:58:57polemonbut more importantly, Opus is in very wide use already
09:59:26polemonthings like Skype, Spotify, etc. streaming services are quite keen on using it
09:59:52[Saint]Indeed. It is incredibly diverse.
10:00
10:00:13*freqmod waits for android support
10:00:23freqmodfor opus
10:00:42[Saint]Well, /technically/ there is Opus support for Android.
10:00:47polemonwell, Google is quite after it, as it will be used in the newer WebM with VP9
10:00:48[Saint]RaaA
10:01:10polemonso Android support shouldn't be too far down the road
10:01:21polemonthen again, Google kinda shunned Linux users with Google Drive
10:01:32*[Saint] does realize that freqmod means "native support"
10:01:38[Saint]I was just being pedantic.
10:01:54freqmodRaaA is nice, but it is not very well integrated
10:02:03freqmodor stable
10:02:12freqmod(when i tried it a few months ago)
10:02:14*polemon has no Android devices
10:11:54 Quit pamaury (Ping timeout: 264 seconds)
10:13:20polemonfreqmod: how did you put Codec2 support into Rockbox?
10:13:44freqmodthe same way i put speex and rockbox support
10:13:51freqmodlinked the library and called it from rockbox
10:14:11freqmodit was just a proof of concept, but i stopped working on it when i didn't see any fixed point version
10:14:23freqmodi am not good enough with dsp to convert it myself
10:17:48freqmodbtw the codec2 format does not support seeking
10:18:08freqmodbut it should be possible to just jump to the correct part of the file and play from there (as it is fixed bitrate)
10:26:57 Join petur [0] (~petur@rockbox/developer/petur)
10:32:58polemonwell, I'd like to see it on Rockbox, because why not?
10:33:19polemondoes the format hint on what kind of compression was used?
10:33:48polemonthe decoder has to be told what compression that is, that he's being fed
10:33:58 Join Norm_ [0] (~Norm@81.213.109.192)
10:42:47 Quit jhMikeS (Ping timeout: 252 seconds)
10:43:29copper07:59:27 UTC <polemon> things like Skype, Spotify, etc. streaming services are quite keen on using it
10:43:40polemonyees?
10:43:42copperit's not implemented in Spotify yet though, right? I only see a feature request.
10:44:04polemonnope, spotify is behind, Skype moved to Opus last month
10:44:23polemonthere is an article about this, but I'd have to look for it
10:45:14copperOpus tools are still missing even on computers
10:45:24polemonVorbis was kinda constrained to embedded stuff and things like computer games
10:45:50polemonyeah, but Opus was RFC approoved just a few months ago
10:46:14polemonthings like Liquidsoap and FFmpeg needs to get the encoders etc
10:46:44 Nick SuperBrainAK is now known as DormantBrain (~andy@74.112.203.135)
10:46:59copperbasic tools such as opusgain and opus tagging are still missing
10:47:43polemonwell, since Opus is contained inside an Ogg container, you use pretty much the same thing like vorbis comment, etc
10:48:21 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
10:48:41copperyes but neither vorbiscomment nor oggz-comment actually support opus
10:48:45polemonopusenc supports that: −−comment, −−artist, −−title
10:48:46coppernot the stable version anyway
10:49:04copperpolemon: that's fine but you need tools to alter tags after encoding
10:49:32polemonyeah, but that's just something like an afterthough
10:49:39polemont
10:49:39polemonot
10:49:50polemonit's more important to get things using it beforehand
10:50:13polemonOpus is best for streaming, so the whole metadata thing, wasn't really that important
10:50:14copperhow can you use it if you don't have the necessary tools
10:50:52polemonyou have the libraries
10:51:15copperusers need tools, not libraries
10:51:29polemonsure, but it's not geared at this kind of "market"
10:51:43polemonVorbis was aimed at the "MP3" market
10:51:50polemonOpus never (and still kinda) isn't
10:51:52copperusers are not the market?
10:52:04copperproducers are users too
10:52:12polemonI've actually talked to Jasper from Xiph the other day
10:52:34polemonit is basically first and foremost for the technical aspect of audio streaming
10:53:31polemonthat Opus may rival things like AAC, was pretty much accidental
10:54:27polemonI was a replacement for Vorbis, but they didn't think it actuall would rival over the whole spectrum
10:54:49polemonit was intended as a low-latency codec, but it turned out to be just as good for archiving
10:55:02polemon(archving meaning having the files for playback)
10:55:15polemonalso, if you're a producer, you'd want the lossless masters
10:55:31polemonthere is no point to keep the Opus files, really
10:55:52polemoneven if you want to stream them, you'd re-encode them with your streaming software into opus from the FLAC masters
10:56:12polemonit's how most webradios work, and that's how FLAC is actually designed to work
10:56:56copperaaaaaah
10:56:59copperhttps://github.com/fmang/opustags
11:00
11:00:24polemonyeah, someone seems to bother with a little tool like that
11:00:29 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:00:47polemonbut there's going to be many more, just like with the plethora of ID3-Tags
11:13:21copperbrb
11:14:59 Quit copper (Quit: ZNC - http://znc.in)
11:16:22 Quit way2strong (Read error: Connection reset by peer)
11:18:35 Join copper [0] (~copper@unaffiliated/copper)
11:20:42 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
11:26:26polemonhmm, opustags
11:26:42polemonthe naming scheme of Xiph is completely going down the gutter
11:26:59polemonit should be oggtags, if anything
11:32:39copper1) the author isn't associated with Xiph
11:32:48copper2) it supports Opus and nothing else
11:33:53copperlike vorbiscomment supports only Ogg Vorbis
11:34:21copperoggz-comment should support Opus too, in the next version
11:36:31 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
11:38:54copperugh, except oggz-comment is a lot less convenient
11:39:08copperit supports opus in git though
11:48:25***Saving seen data "./dancer.seen"
11:58:19polemoncopper: except it's the same thing in all of them
11:58:47polemonvorbis comment and opus comment is actually the same thing
11:59:00polemonor another way would be to say, Opus uses vorbis comments
11:59:14copperexcept that oggz-comment needed to be modified in order to support it
11:59:15polemonhttp://en.wikipedia.org/wiki/Vorbis_comment <−−- first line of text
11:59:23copperso it's not all that universal
11:59:57polemonit's just down to the tools not understanding the universality of it
12:00
12:00:02polemonbut it actually is universal
12:00:08copperoggz-comment is from Xiph itself
12:00:13copperso if THEY don't understand it…
12:00:26coppersomeone failed, somewhere.
12:00:34polemonin fact Xiph have some of their "convention" problems in many places
12:01:04polemonfor one thing, the opus tools don't behave like any other tool for Xiph codecs
12:01:30polemonno short options and up to the very recent SVN builds, opusenc couldn't decode FLAC files
12:02:26polemonit's things like that, then of course extensions: .flac <−−- whether it's inside an Ogg container or not, is not determinable by extension
12:02:48polemon.ogg <−−- that's the extension of Ogg/Vorbis, but .opus is the extention of Ogg/Opus...
12:03:40polemonthey neglected OggPCM to the point where it is in the spec, but not used anywhere and by none of their tools
12:03:47copperOgg FLAC extension is .oga
12:03:57copperor .ogg
12:04:00polemonit would be a great stream codec for piping things into one another
12:04:18polemoncopper: .flac is a valid Ogg/Flac extension, though
12:05:02copperthe flac encoder sets it to .oga by default
12:05:13copperand the man page says to use .oga, too
12:05:16[Saint]It is, yes, but people that use that need to be shot in the face.
12:05:30polemonhehe
12:05:30[Saint]FLAC in an ogg container == .oga
12:05:47[Saint]...and, I see I'm late to the party and copper already said that.
12:05:49polemonwhy not vorbis in an ogg container = .oga
12:05:55polemonwould make sense, wouldn't it...
12:06:24copperah
12:06:41coppermetaflac can't deal with .oga files!
12:07:17*[Saint] thinks this should /probably/ shift into *-community now
12:07:24copperand neither can oggz-comment
12:07:25polemonsuffice it to say, when it comes to naming and operations conventions, Xiph is unfortunately a big chaotic hairball
12:07:38copperhow's that for "universal"
12:23:55coppermy bad, oggz-comment does read tags from FLAC .oga files
12:43:13 Part Norm_ ("Leaving")
13:00
13:09:08 Join Guest29983 [0] (~liar@clnet-p09-185.ikbnet.co.at)
13:10:38 Join kilroy [0] (~dewlap@2001:0:5ef5:79fd:2c93:e83:b815:38b6)
13:12:58 Join traps [0] (~dewlap@2001:5c0:1000:a::1dff)
13:13:27 Quit dewlap (Ping timeout: 245 seconds)
13:15:57 Quit kilroy (Ping timeout: 245 seconds)
13:38:09 Quit wodz (Ping timeout: 240 seconds)
13:38:31 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
13:48:02 Join way2strong [0] (~irssi@pool-173-79-215-215.washdc.fios.verizon.net)
13:48:27***Saving seen data "./dancer.seen"
13:50:39 Join dfkt [0] (OxO29A@unaffiliated/dfkt)
13:53:42 Join petur2 [0] (~petur@91.183.48.77)
13:55:36 Quit petur (Ping timeout: 256 seconds)
13:56:26 Nick petur2 is now known as petur (~petur@91.183.48.77)
13:56:36 Quit petur (Changing host)
13:56:36 Join petur [0] (~petur@rockbox/developer/petur)
14:00
14:48:55 Part LinusN
15:00
15:32:53 Quit mortalis (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
15:48:28***Saving seen data "./dancer.seen"
16:00
16:01:41 Join amayer [0] (~amayer@mail.weberadvertising.com)
16:11:35 Join krabador [0] (~krabador@unaffiliated/krabador)
16:37:49 Quit melmothX (Quit: #)
16:49:17 Join redhot [0] (b0f16b6a@gateway/web/freenode/ip.176.241.107.106)
16:49:22redhotHowdy!
16:49:47redhotDoes anyone faced and issue with SanDisk Sansa Clip+ and 32GB microSD card?
16:50:08redhotMy card loses some files accidentaly, and it happens from time to time
16:50:36redhotmicroSD is SanDisk Ultra 32GB
16:50:56redhotRockBox is 3.13
16:52:20 Quit olspookishmagus (Quit: free() the malloc())
16:53:46copperredhot: Nope. When does it happen?
16:56:13redhotAccidentally... Few times. I switch on player and see some folders are completely empty
16:56:49redhotRe-checked on PC and it seems like these files just dissappeared. From few different folders
16:57:05copperwhere they ever there to begin with?
16:57:29redhotI have put files from PC to microSD and they were OK for few days
16:57:45redhotThat after few player restarts they started to disappear
16:57:54redhotI suspect it may be a card issue
16:58:04redhotBut need to clarify
16:58:25copperthat's never happened to me
16:58:43copperbut I do believe that the quality of the card play a role in overall stability
16:58:45redhotI know, I'm lucky to get enough exceptions
16:58:47copperplays*
16:59:10KohlrabiSounds like your card has problems
16:59:39redhotSo basically size (32GB) is OK for Rockbox and Player , yep?
16:59:44Kohlrabiyes
16:59:56redhotIt should be card quality issue?
17:00
17:00:02KohlrabiSounds like it
17:00:26KohlrabiDid your reported volume size decrease on that card?
17:00:37copperKohlrabi: uh?
17:00:43redhotyes
17:00:46Kohlrabicopper: Disabled sectors
17:01:01copperhow does it affect volume?
17:01:04redhotWait, what Soft to check?
17:01:30Kohlrabicopper: Disabled sectors should decrease available size
17:01:40copperoooh
17:01:54redhotKohlrabi: and disabled sectors means "bad" sectors, huh?
17:01:55copperI missed the "size" part
17:01:58KohlrabiYes
17:02:03copperI thought you were talking about sound volume
17:02:16KohlrabiI think flash card controllers are able to tag broken sectors and disable them
17:02:23redhotyay! It is definitely so.. What can I use to check "disabled" sectors size?
17:02:27copperdoesn't the filesystem do that?
17:02:33Kohlrabicopper: Does the FS do that?
17:02:41KohlrabiI thought it was down on Firmware or hardware level
17:02:53Kohlrabiredhot: Dunno, google :=
17:02:54Kohlrabi:)*
17:03:03redhotit's ok :)
17:03:56copperredhot: don't be cheap on the microsdhc card, buy a high quality one from a reputable source, it will save you lots of headaches
17:04:28copperKohlrabi: badblocks I guess (on linux / perhaps unix)
17:05:01copperso yeah, my bad, I don't think filesystems do that
17:05:02redhotcopper, huh this is expensive and genuine SanDisk UltraSDHC
17:05:11copperredhot: where did you buy it?
17:05:19redhothuh, ebay)))
17:05:31redhotbut Sandisk confirms it's genuine
17:05:35redhotby S/N
17:05:50coppercan't the S/N be simply copied (faked)?
17:06:09copperthe faker buys a genuine card, copies the S/N, done
17:06:36redhotcopper: not in this case, I've checked a lot of things and it sands like Genuine
17:06:44redhotI've bought a fake one before :)
17:06:51redhotSo I could compare
17:06:54redhot:D
17:06:57copperwell I bought a seemingly "genuine" transcend card that gave me problems right from the start
17:07:07copperexcept I bought it from a chinese hardware store
17:07:31copperI trust neither chinese stores nor eBay
17:07:48redhotcopper: Yep, thanks guys! Seems like I'll contact Sandisk concerning this item again
17:08:18copperredhot: I bought a new SanDisk card from Amazon and my problems magically disappeared
17:08:32copperJust sayin'.
17:09:08redhotcopper: Yep, I've also bought a microSD from Amazone and it works just well :)
17:10:17copperdoes your faulty card look like this? http://www.sandisk.com/assets/Product/microsd-sdsdqua-032g.jpg
17:11:00redhotSure, it has qulity materials and printing
17:12:51copperredhot: why operating system do you use?
17:14:09redhotcopper: Why? Because I dont like bare metal things without OS :)
17:14:23redhotActually I used WinXp only to put files
17:14:32redhotThen used card in Clip+ only
17:14:38copperif you run Linux, I have a command that you can run to test the card's speed
17:15:06redhotyep, I know about dd, thanks. Speed is crappy too :)
17:15:12copperhow crappy
17:15:31redhotSandsik support says it's ok) It is ~6-8 MB/s in real world :)
17:15:40copperuh
17:15:45copper6-8MB/s is not crappy
17:15:49redhotOn benchmarks it is ALMOST 10 mb
17:15:53redhotthat's class 10
17:16:14redhotfor class 4 or 6 it'll be ok
17:17:17copperthey liz
17:17:19copperlie
17:17:47copperbut given the speed you're reporting, I'd wager that the card is genuine but faulty
17:17:49redhotFor comparison I've bought Transcent Ultra card (16GB) - it gives me real 10-12MB/s...
17:17:58coppermicrosdhc too?
17:18:01redhotyep
17:18:04redhotbut 16GB
17:19:30redhotI thinks I will contact theirs Support. It is definitely faulty... Who know's maybe after weeks of emailing (LOL) they will replace it :)
17:19:43copperamazon would do it in a blink
17:20:09copperalso, I'm not sure SanDisk will comply, given that you bought it on eBay
17:20:34redhotcopper: SanDisk once replaced my broken Clip+ :)
17:20:40copperthat's exactly the kind of thing you should NOT buy on eBay
17:20:46redhotSo I've got a good opinion about them
17:21:44redhot"that's exactly the kind of thing you should NOT buy on eBay" - it's a kinda lottery, you are right
17:22:16redhotok, guys thanks a lot for your help :-)
17:23:39 Quit kevku (Ping timeout: 260 seconds)
17:26:18redhotHave a nice day everyone! ;-)
17:26:50 Quit redhot ()
17:38:39 Quit petur (Quit: Nettalk6 - www.ntalk.de)
17:42:26 Quit wodz (Read error: Connection reset by peer)
17:47:25 Join bady48 [0] (~594a2b27@www.haxx.se)
17:48:06 Quit bady48 (Client Quit)
17:48:29***Saving seen data "./dancer.seen"
18:00
18:07:08 Join Raptors_ [0] (~whoneedsa@216-58-33-203.cpe.distributel.net)
18:07:08 Quit Raptors (Disconnected by services)
18:08:29 Quit polemon (Ping timeout: 256 seconds)
18:22:56 Quit Raptors_ (Ping timeout: 276 seconds)
18:29:53 Join jokke [0] (~jokke@ip-130-180-2-169.unitymediagroup.de)
18:30:39jokkehi! anyone here playing lots of opus encoded music on his/her rockbox device? I'm curious about the battery consumption.
18:37:14 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:40:34 Join TeruFSX2 [0] (~TeruFSX@75-163-26-8.dlth.qwest.net)
18:43:45TeruFSX2i was looking at the issue with ipods and chargers and i thought one possible option would be a timeout
18:45:02TeruFSX2you would turn this option on, and if the other device looks like it's been resetting the bus for a second, you assume it's not actually a live device and back out of disk mode
18:46:54 Join belak [0] (~belak@facebook/engineering/belak)
18:49:01 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
18:59:43 Join Raptors [0] (~whoneedsa@216-58-33-203.cpe.distributel.net)
19:00
19:03:12 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
19:20:49 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
19:21:16 Join jlbiasini [0] (~metaphysi@86.34.67.25)
19:22:09jlbiasinipamaury: bluebrother: this one is for you... g#495
19:22:12fs-bluebotGerrit review #495 at http://gerrit.rockbox.org/r/495 : [Manual] Document soft lock for FM by Jean-Louis Biasini (changes/95/495/1)
19:22:49jlbiasiniplain right copy/paste from WPS
19:23:10jlbiasiniStill I suppose it makes sense adding
19:23:46 Join wodz [0] (~wodz@89-75-41-78.dynamic.chello.pl)
19:23:55wodzpamaury: ping
19:27:16pamaurywodz: pong
19:28:08copperTie.
19:28:45wodzpamaury: Could you run some code on your rk27xx dap? I created hacky patch to dump begining of the nand to the file. Maybe this will proof my idea why you only have system partition
19:29:44pamaurywodz: ok, send me a mail and i'll do it at home, I need to go back home now
19:30:13wodzok
19:32:29 Join n1s [0] (~n1s@nl118-168-30.student.uu.se)
19:32:29 Quit n1s (Changing host)
19:32:29 Join n1s [0] (~n1s@rockbox/developer/n1s)
19:32:47wodzpamaury: you should have test build + instruction in your mailbox
19:35:27 Quit pamaury (Ping timeout: 248 seconds)
19:43:27 Nick DormantBrain is now known as SuperBrainAK (~andy@74.112.203.135)
19:48:30***Saving seen data "./dancer.seen"
20:00
20:03:45 Quit Guest29983 (Ping timeout: 268 seconds)
20:06:33amayeris anyone else having problems with the unread forum posts page? it says i dont have any unread posts, but the homepage shows blue squares all over the place
20:11:07jlbiasinihi all: anybody know why all screen pictures are blank when I build the manual?
20:12:01jlbiasiniit seems only the border of the image get displayed...
20:12:55 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:12:58jlbiasiniha ok never mind it's the soft I was using that doesn't read the pdf correctly
20:43:00 Quit tchan (Quit: WeeChat 0.4.1)
20:43:20 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
20:43:36 Quit Scall (Ping timeout: 256 seconds)
20:46:45 Join Scall [0] (~chat@unaffiliated/scall)
20:48:58 Quit jlbiasini (Quit: jlbiasini)
20:49:28 Quit pamaury (Read error: Operation timed out)
20:51:08 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:51:51 Quit Scall (Ping timeout: 252 seconds)
20:54:15 Join Scall [0] (~chat@unaffiliated/scall)
21:00
21:02:29 Quit Zarggg (Read error: Connection reset by peer)
21:03:05 Join Zarggg [0] (~zarggg@24.229.140.62.res-cmts.sm.ptd.net)
21:19:41 Quit TeruFSX2 (Ping timeout: 264 seconds)
21:28:40 Join TeruFSX2 [0] (~TeruFSX@75-163-26-8.dlth.qwest.net)
21:42:37pamaurywodz: see mailbox
21:43:04wodzpamaury
21:43:14wodzok
21:44:31wodzpamaury: hmm, id block is set for 60MB system partition
21:48:32***Saving seen data "./dancer.seen"
21:55:27 Quit Rower (Quit: Hmmm...)
22:00
22:11:07 Join vis_ionmas_ter [0] (4cb1c293@gateway/web/freenode/ip.76.177.194.147)
22:11:30 Part vis_ionmas_ter
22:12:29n1scould someone who understands make review g#496 ?
22:12:32 Quit fs-bluebot (Remote host closed the connection)
22:13:14 Quit y4n (Quit: MOTHER EUROPA CALLING ME!)
22:13:39wodzn1s: you basically call gevaerts :-)
22:13:59gevaertsSorry, no. Variables are a part of make I don't get :)
22:14:37gevaertsMaybe Zagor?
22:17:41n1shmm, in fact this might be doable with a much smaller change. The recursively expanded vars really are confusing
22:24:55 Quit krabador (Quit: Bah...)
22:27:35 Quit n1s (Quit: Ex-Chat)
22:34:14 Join saratoga [0] (d8507479@gateway/web/freenode/ip.216.80.116.121)
22:38:51 Join jhMikeS [0] (~jethead71@d192-24-174-117.try.wideopenwest.com)
22:38:51 Quit jhMikeS (Changing host)
22:38:51 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
22:44:33 Quit GeekShadow (Ping timeout: 246 seconds)
22:45:45 Join GeekShadow [0] (~antoine@reactos/tester/GeekShadow)
23:00
23:03:34 Nick SuperBrainAK is now known as DormantBrain (~andy@74.112.203.135)
23:03:55 Nick DormantBrain is now known as SuperBrainAK (~andy@74.112.203.135)
23:13:51 Quit kevku (Ping timeout: 245 seconds)
23:15:02 Quit amayer (Quit: Leaving)
23:24:27 Quit makoto (Ping timeout: 246 seconds)
23:33:08 Join makoto [0] (~makoto@hey.dont.privmsg.me.uk)
23:34:44 Quit ender` (Quit: You must know by now we Brits are going metric inch by inch! -- Greg Chapman)
23:43:44 Join prof_wolfff [0] (~prof_wolf@62.83.50.196.dyn.user.ono.com)
23:48:35***Saving seen data "./dancer.seen"
23:49:54 Join dewlap [0] (~dewlap@2001:5c0:1000:a::1dff)
23:52:36 Quit traps (Ping timeout: 245 seconds)

Previous day | Next day