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 2013-11-22

00:00:12coppergood night
00:00:15[Saint]ANd it is possible to remove the Apple firmware, and it will actually increase the boot time by a fairly considerable amount doing so.
00:00:23[Saint]I'll get you the link. One second.
00:00:36GLC-ok so the rockbox system/os then finds music files and arranges them to some kind of library?
00:01:26[Saint]Rockbox has a file borwser, and a database that uses track metadata.
00:01:48[Saint] <−− here is the link you would need to completely remove the Apple firmware, should you want to.
00:01:49wodz[Saint]: you mean decrease boot time, right?
00:02:02[Saint]wodz: errr, yes. I do. Whoops.
00:02:41*gevaerts recommends going with the standard installation method at first
00:02:42GLC-but wouldn't it be good to get rid of apple? i didn't bought ipod back in the day because it was Apple or ipod, it was slim sleek nice design 30GB MP3 player affordable price. so this Rockbox thing really sounds good, because i wanna sync with WMP12
00:02:42[Saint]From that link, I use option: 4) OSOS contains only Rockbox
00:03:15GLC-standard? but it installs radio too? there is no radio.. in that model
00:03:30GLC-thanks for the link Saint
00:03:37[Saint]There is no real reason to remove the Apple firmware, ROckbox will boot by default anyway and you still have the option of booting the Apple firmware should you chose to.
00:04:08[Saint]But if you know you will never use the Apple firmware, there's no real point in keeping it there either I guess.
00:04:32 Quit ender` (Quit: The process takes just over six seconds, and would sound like a coffee machine passing an aluminum baseball bat if there were any atmosphere to carry the sound. -- Howard Tyler: Schlock Mercenary)
00:04:34 Quit petur (Quit: Leaving)
00:05:05gevaertsGLC-: there is a radio in some accessories
00:05:05GLC-yeah I dunno, can I fuck up the drive? Couldn't I just reload the original firmware with iTunes?
00:05:33GLC-gevaerts, ah ok, good to know, those must be going cheap these days
00:06:11gevaertsAnd yes, you can always easily restore with itunes
00:06:15[Saint]Yes, you can always restore the device with iTunes should you want to.
00:06:17GLC-hmhmh maybe I'll leave the apple stuff there and decide later
00:06:41GLC-thanks both of you
00:06:42[Saint]That is a wise choice.
00:06:44 Quit kugel (Ping timeout: 248 seconds)
00:06:56gevaertsThat generation of ipods is really hard to break in a way that's hard to recover, unless you use the hammerâ„¢ method
00:07:12GLC-so with rockbox i can also use ipod sort of as giant usb stick then?
00:07:30[Saint]It pays to note that if you do use any accessories it would be wise to keep the original firmware and use an Application that is capable of syncing to the iTunes Database.
00:07:39GLC-yeah, it's the iPod video (5G) model
00:08:17GLC-yeah ok but there's no accessories in sight
00:08:47GLC-if i would of known all this Rockbox stuff 7 years ago.. :)
00:09:48wodzpamaury: Updated gerrit. I wonder if the same fix in rb driver will make usb failures fixed...
00:09:51GLC-does Rockbox increase battery life? compared to icrap?
00:10:47pamaurywodz: that's a good question indeed
00:10:55GLC-i mean does rockbox consume less energy than ipodsoft?
00:11:02[Saint]It can. But it can also allow you to set a terrible configuration that would hit the battery heavily.
00:11:30[Saint]It depends on the usage really.
00:11:51gevaertsIt's likely going to be more or less the same
00:11:51GLC-ah ok, thanks, yeah I'm only going to listen music despite all the video etc, so yeah I should keep install minimum
00:12:12gevaertsapple does a fairly good job at battery usage in their firmware usually
00:12:31GLC-hmm, ok
00:12:51[Saint]They are definitely a lot better at video on this device than Rockbox is.
00:13:03gevaertsDon't compare to what you remember from seven years ago though, the battery isn't going to be as good any more :)
00:13:06GLC-but yey! now I can sync with WMP12 how fantastic is this?!!!1 thank you guys
00:13:15 Quit krnlyng (Remote host closed the connection)
00:13:55GLC-true, but it's not half life yet either
00:14:17[Saint]Album art should "Just Work" if you sync with WMP as well, which is a bonus I guess.
00:14:43GLC-yeah, only music and perhaps an occasional ebook if I go away for awhile, like vacation etc
00:15:00GLC-great! yeah album art is a must have
00:15:05[Saint]I do recommend reading through our fine manual:
00:15:44GLC-thanks, I'll read the links before plugging the pod
00:15:52GLC-thanks for the link
00:16:14pamaurydamn, the freescale nand code is uber-complicated
00:16:34pamauryi'm trying to reverse the FTL but I'm still far from done with the pure nand layer yet
00:17:12GLC-yeah, I have plenty to do before i get to conf my ipod, in a few days, I'll report here how it goes, my first imressions etc (not that I think anynone's interested but i like to think loud)
00:17:41[Saint]Installation is really a very straightforward process.
00:18:05[Saint]ANd it is pretty much impossible to hurt the iPod in an irrecoverable way.
00:18:45GLC-yes I saw some screenshots of the installer, looked easy, I'm not novice on PC
00:18:45wodzpamaury: In case of bulk ep this should be the same, if so how it was ever even half-working. I suspect lots of multi 512b transfers are used.
00:18:49[Saint]Installation literally involves clicking a few buttons and waiting about a minute, depending on your internet connection.
00:19:27GLC-and hey it's goof to know that the pod cannot brake down, yey! :)
00:20:21[Saint]It certainly is possible to do some damage with our tools, but you would need to be trying *really* hard to do so - it would need to be a deliberate effort.
00:20:55pamaurywodz: indeed, luck maybe ^^
00:21:10[Saint]Using things as intended is pretty much bullet proof though.
00:21:11pamauryrockbox doesn't use many blocking transfers
00:22:30GLC-ok, yeah, just click few buttons on the installer, shouldn't be that difficult
00:23:04GLC-oh hey it has a lot of music in it so I guess I'll have to format it before rockboxing?
00:23:30GLC-it's probably in the manual :)
00:25:00[Saint]If you have the music backed up elsewhere, it is probably best to do a restore with iTunes just before you install Rockbox.
00:25:18[Saint]You can format it manually, but iTunes is absolutely guaranteed to get it right.
00:26:22[Saint]The other option is mounting the device in disk mode, enabling "show hidden files", and just deleting everything on the disk.
00:26:37[Saint]Formatting the disk is a bit harsh.
00:44:46 Quit lebellium (Quit: ChatZilla [Firefox 26.0/20131118212339])
00:55:10 Quit zoktar (Ping timeout: 245 seconds)
00:55:51 Quit wodz (Quit: Leaving)
01:05:48 Join zoktar [0] (~zoktar@unaffiliated/zoktar)
01:19:39 Quit pamaury (Ping timeout: 264 seconds)
01:34:03 Quit cmhobbs (Ping timeout: 265 seconds)
01:38:12***Saving seen data "./dancer.seen"
01:41:53 Join treaki_ [0] (
01:44:20 Quit treaki__ (Ping timeout: 248 seconds)
01:52:33 Join cmhobbs [0] (
01:52:33 Quit cmhobbs (Changing host)
01:52:33 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
02:38:38 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
02:56:22 Nick DormantBrain is now known as SuperBrainAk (~andy@2001:470:8:a61::5f92:59a1)
03:15:56 Quit cmhobbs (Ping timeout: 272 seconds)
03:36:06 Join Strife89 [0] (
03:38:16***Saving seen data "./dancer.seen"
04:09:39 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
04:10:23 Join krabador [0] (~krabador_@unaffiliated/krabador)
04:18:06 Quit cmhobbs (Ping timeout: 272 seconds)
04:18:26 Join linuxguy3 [0] (
04:23:56 Quit mc2739 (Quit: leaving)
04:30:35 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
04:33:23 Quit [Saint] (Remote host closed the connection)
04:33:43 Quit amiconn (Disconnected by services)
04:33:43 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:33:46 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:34:21 Quit pixelma (Read error: Operation timed out)
04:34:47 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
04:35:45 Join [Saint] [0] (~saint@rockbox/user/saint)
04:37:06 Quit [Saint] (Remote host closed the connection)
04:39:10 Join [Saint] [0] (~saint@rockbox/user/saint)
04:39:36 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
04:46:39 Quit cmhobbs (Ping timeout: 264 seconds)
04:48:39 Quit krabador (Quit: Sto andando via)
05:08:26 Quit TheSeven (Read error: Operation timed out)
05:10:38 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:38:19***Saving seen data "./dancer.seen"
06:17:25 Quit foolsh (Remote host closed the connection)
06:18:57 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
06:21:41DuperManipod mini witf cf == sux in rockbox, great in of
06:21:49DuperManjust got a sansa clip+
06:22:03DuperMansweet little beast but suchhhh tiny screen^_^
06:24:17 Quit cmhobbs (Ping timeout: 245 seconds)
06:38:50JdGordonDuperMan: you've been told before, use full english words only
06:48:24*DuperMan checks if his build for the clip+ works
06:48:49DuperManI do, JdGordon. some people could parse better, but that's no fault of mine.
06:49:51JdGordonNo room for discussion, this channel is logged and viewed by non-English speakers. It's up to you if you want to play by the rules and stay
06:50:44lonoxmontlisten to them man, its pretty strict here
06:50:57lonoxmontand with good reason
06:51:00DuperManI am playing by the rules. you're both o/t, btw. :)
06:51:05lonoxmonthard to autotranslate gibberish
06:52:07DuperManoff topic. back on topic, my build for the scp is working well, unsurprisingly since I merely compiled from the main branch
06:53:24DuperManohhh... 'witf' meant with, sorry for the typo if that's what confused you Jd
07:01:55 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
07:02:58 Quit dfkt (Ping timeout: 252 seconds)
07:06:11 Join thegeek_ [0] (
07:06:19 Quit thegeek (Read error: Connection reset by peer)
07:06:19 Quit Jinx (Read error: Connection reset by peer)
07:38:21***Saving seen data "./dancer.seen"
07:53:27 Join mortalis [0] (~kvirc@
08:09:54 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:19:18 Nick SuperBrainAk is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
08:26:43 Join Zagor [0] (
08:26:43 Quit Zagor (Changing host)
08:26:43 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:46:50 Quit bertrik (Remote host closed the connection)
08:52:35copperBattery benchmark of the iPod Classic, -24 dB, lossyFLAC, with a build from 2012-11-03:
08:52:59copper30h 37m, compared to 25h 38m with the latest build:
08:53:09copperthere is definitely a battery life regression here
08:53:37coppera rather big one
08:54:14copperTheSeven: ^^
09:07:58 Join wodz [0] (
09:09:44 Quit kevku (Ping timeout: 245 seconds)
09:27:54 Join LinusN [0] (
09:32:54 Join lebellium [0] (
09:33:57 Join ender` [0] (
09:37:52copperhopefully you guys will be able to make an educated guess as to what commit may have caused it, because I haven't the faintest clue
09:38:23***Saving seen data "./dancer.seen"
09:39:05copperalso, why doesn't git log display commits ordered by date?
09:39:49copperlike, I see commits from Nov 18, then Nov 9, and Nov 15, in that order
09:39:54copperit's super confusing
09:42:37 Join petur [0] (5bb7304d@rockbox/developer/petur)
09:43:12DuperManit displays ordered by latest comment iirc
09:43:15amiconnThe commits are ordered the way they were pushed. The commit dates reflect when the author committed them locally
09:43:18DuperManand not really
09:43:31DuperMan(not confusing)
09:49:11wodz[Saint]: Maybe you could make battery_bench on n2g to check if we have regression there as well? According to [7] n2g can measure current draw using internal adc so this would allow to fast(er) pinpoint offending commit
09:51:07copperamiconn: I see, thanks
09:53:27 Quit mortalis (Quit: KVIrc 4.3.1 Aria
09:54:15 Join mortalis [0] (~kvirc@
09:56:46copperOn July 28 (2013), I was asked to run some codec benchmarks (test_codec). I don't know if anyone later committed stuff related to that, and whether it could have caused the regression.
09:58:35copperI also remember related talk about "boosting"
09:59:59 Join kevku [0] (~kevku@2a01:d0:ffff:34a::8:3)
10:04:36GodEatercopper: are you aware of "git bisect" ?
10:04:46 Join krnlyng [0] (~liar@
10:04:59copperbut I'm not going to bisect a year's worth of commits
10:05:14GodEaterso you want someone else to?
10:05:33copperit's not that
10:05:48copperI figure you guys are a lot more likely to be able to make educated guesses
10:06:26GodEateryou'd be amazed at the number of regressions which defy educated guesses
10:09:03GodEaterit's not the bisecting that'll take the time though in this case, it's having to run a battery bench with each build
10:09:14copperthat too
10:09:29GodEaterwhich, since you're the one with the issue - you're going to have to do anyway
10:09:38GodEaterso you might as well make a start on the bisecting :)
10:09:44copper"I'm the one"?
10:10:33copperI highly doubt this is a freak occurence that only affects me
10:10:40GodEaterunless I've missed the hordes of other people reporting it?
10:11:28coppera difference between 30 hours and 25 hours might not be obvious when using the iPod normally, for listening to music
10:11:56copperand the difference could easily be attributed to an aging battery
10:12:26copperand, few people ever take the time to report bugs and run battery benchmarks
10:13:05wodzcopper: Do you have battery_bench of old build made currently to exclude battery aging & co ?
10:13:25copperI'm sorry, I don't know what you mean
10:13:54copperI ran both battery benchmarks in the last couple of days
10:14:29copperwith a build from this month, and then a build of a year old commit
10:15:03wodzcopper: that what I mean
10:15:11copperthe build dates in the logs I posted reflect that
10:19:35DuperManlogs should have more vaudeville
10:19:41DuperManfor mass appeal
10:19:59gevaertsDuperMan: stop it *now*
10:20:26marazofftopic/social chat -> #rockbox-community
10:20:46DuperMandone. by the by, I finally set up a working build env so I might become productive yet :) DuperMan out
10:51:22 Quit Strife89 (Ping timeout: 245 seconds)
11:05:49copperthe Fuze+ may be affected too, unless it's just a battery calibration thing: it's currently at 55% battery, after over 16 hours of playback, versus 12 hours 30 minutes with the recent build
11:06:09copperbut that's what I saw with the Classic too
11:07:39copper(pamaury asked me to benchmark an old revision from last July)
11:08:29copperinterestingly, I've been benchmarking different codecs on the Classic and the Fuze+: FLAC on the Classic, MP3 on the Fuze+
11:08:37copper(lossyFLAC and LAME V0)
11:10:40copperit should complete late this night, I'll have the result tomorrow morning
11:33:43 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:34:09pamaurycopper: on my fuze+ with an old revision, I am currently at 19:13 with battery at 3762
11:34:39pamaurywhich is very comparable to HEAD
11:35:20pamaurythe reason to choose those commits is that I think HEAD and january13 are at the same level but may13 was a few hours better
11:35:28copperc4f2a46e0dfee336ce7016cd62608097f15367b8 ?
11:37:29pamauryyeah, well it's more july13 actually ^^
11:38:16copperso, no relation to the Classic battery life differences?
11:38:25***Saving seen data "./dancer.seen"
11:39:00pamauryI don't expect so, the problem I'm chasing is imx233/fuze+ specific
11:39:03copper(nov 2012: 30 hours, head: 25 hours)
11:39:16pamaurybut many something is affecting both
11:39:22pamaurywho knows ^^
11:40:03pamaurythat's the point of benchmarking, trying to find out
11:40:51pamaurywodz: did you try the ZLP fix with rockbox usb driver ?
11:41:58wodzpamaury: yes, but I was very tired and changed init sequence a bit which resulted in not enumerating at all
11:43:10wodzmaybe tonight I'll find time to check changes one-by-one
11:54:15pamauryTheSeven: ping
11:55:19DuperManpamaury: finally set a build env. I am having trouble figuring out how to 'mess' with the plugins but will figure it out eventually:)
11:56:05pamauryDuperMan: you have two ways:
11:56:05pamaury1) compile with an additional define: "make ENABLEDPLUGINS=yes"
11:56:31pamaury2) enable plugins in configure: in tools/configure, find the "creativezen)" section and change plugins="" to plugins="yes"
11:58:19DuperMantried, failed. it built like 2 or 3 plugins. can't go deeply into it but thanks, I'll be back in the evening (I hope)
12:00:05pamauryI suggest 2) so you don't have to remember the magic define all the time
12:00:07pamauryyou will get many compile errors, that's normal
12:00:08pamauryyou need to go through each plugin which fails and add a keymap
12:02:06DuperManahhh... guess I'd love to do so if I'll have the time tomorrow. might need some handholding but hey - I'm doing the menial labor here ;) DuperMan out/bbl
12:02:52 Quit DuperMan (Remote host closed the connection)
12:52:27 Join mudlord [0] (
12:53:40 Join krabador [0] (~krabador_@unaffiliated/krabador)
13:08:10 Quit krabador (Ping timeout: 260 seconds)
13:23:56 Quit Zambezi (Ping timeout: 272 seconds)
13:29:11 Join krabador [0] (~krabador_@unaffiliated/krabador)
13:35:56 Join markun [0] (~markun@rockbox/developer/markun)
13:38:27***Saving seen data "./dancer.seen"
13:39:49 Join Zambezi [0] (
13:39:49 Quit Zambezi (Changing host)
13:39:49 Join Zambezi [0] (Zulu@unaffiliated/zambezi)
13:41:54 Quit DexterLB (Read error: Connection reset by peer)
13:46:56 Join DexterLB [0] (~dex@
13:48:28 Quit krnlyng (Remote host closed the connection)
14:25:02 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
14:30:26 Quit cmhobbs (Ping timeout: 272 seconds)
14:31:20 Join cmhobbs [0] (
14:31:20 Quit cmhobbs (Changing host)
14:31:21 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
14:45:48 Quit cmhobbs (Read error: Operation timed out)
15:09:18 Join amayer [0] (
15:11:47 Quit krabador (Quit: Sto andando via)
15:15:14 Quit wodz (Quit: Leaving)
15:19:48 Quit linuxguy3 (Ping timeout: 248 seconds)
15:20:46 Join krabador [0] (~krabador_@unaffiliated/krabador)
15:25:16copperyay, my 2 band parametric EQ sounds exactly like I want
15:25:35coppergevaerts: I feel less stupid now
15:27:39 Quit kevku (Ping timeout: 245 seconds)
15:35:19 Quit Zagor (Quit: Clint excited)
15:37:57copperand unsurprisingly, it sounds just like my desktop / USB DAC
15:38:28***Saving seen data "./dancer.seen"
15:40:03*copper wonders if anyone uses a Rockbox simulator as a desktop player
16:01:24 Join ikeboy [0] (
16:01:32 Quit krabador (Quit: Sto andando via)
16:11:51 Join W0rmDr1nk [0] (~wormdrink@unaffiliated/wormdrink)
16:12:02W0rmDr1nkhas anybody ever fiddled with transcend MP870 ?
16:14:27 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
16:15:25 Join krabador [0] (~krabador_@unaffiliated/krabador)
16:27:13 Quit lebellium (Ping timeout: 265 seconds)
16:27:14 Join lebellium_ [0] (
16:27:26 Nick lebellium_ is now known as lebellium (
16:29:51 Quit ikeboy (Ping timeout: 246 seconds)
16:32:51 Quit petur (Ping timeout: 250 seconds)
16:35:42 Join Narod [0] (
16:36:14 Part LinusN
16:47:49 Quit krabador (Quit: Sto andando via)
16:52:55 Join einhirn [0] (
16:54:18 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
17:00:37 Quit Narod ()
17:04:35 Join Strife89 [0] (~Strife89@2602:306:250e:659:7d7b:2850:4d11:8c29)
17:05:06 Join cmhobbs [0] (
17:05:06 Quit cmhobbs (Changing host)
17:05:06 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
17:07:52 Join n1s [0] (
17:07:52 Quit n1s (Changing host)
17:07:52 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:16:41 Quit markun (Read error: Operation timed out)
17:16:58 Join rela [0] (~x@pdpc/supporter/active/rela)
17:27:02 Join saratoga [0] (123e11e0@gateway/web/freenode/ip.
17:27:22saratogais there any way we could set the front page to list the main rockbox git commit time, rather than the date the patch was posted to gerrit?
17:27:40saratogait'd be more useful to know when a build began than when somethign went on gerrit
17:38:29***Saving seen data "./dancer.seen"
17:44:16 Join toffe82 [0] (
17:55:22 Quit mortalis (Ping timeout: 246 seconds)
17:56:46 Quit pamaury (Ping timeout: 245 seconds)
17:57:21 Join ikeboy [0] (
18:00:22 Join lorenzo92 [0] (~chatzilla@
18:02:06 Join krabador [0] (~krabador_@unaffiliated/krabador)
18:02:49 Quit ikeboy (Remote host closed the connection)
18:04:20lorenzo92pamaury: I'm super interested in your work for NAND...if you need some help, we could split some disassembling work ^^
18:04:45lorenzo92because if we end up writing a good FTL we might use it also on Z5
18:07:53lorenzo92I forgot to say that, in case, we have the sources for samsung FTL ^^
18:09:59lorenzo92has anyone saw this project?
18:13:16lorenzo92* seen :D
18:15:51 Join foolsh [0] (
18:21:13 Quit lorenzo92 (Ping timeout: 245 seconds)
18:26:32copper[Saint]: instead of blindly implementing iTunes EQ presets, it might be worthwhile to figure out what kind of presets are of general interest
18:27:01copperisn't a V shaped EQ pretty much what most people using presets want?
18:27:28copperit would be nice to have some stats
18:27:42 Join lorenzo92 [0] (
18:27:54 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:27:54copperthat one, for instance, looks silly
18:28:11copperno bass, no treble, all mids, for "pop" music??
18:28:56copper(though that one couldn't be easier to implement with the Rockbox parametric EQ)
18:29:40coppersurely that's the reason even Rockbox has fool-proof "bass" and "treble" settings
18:32:14lorenzo92regarding the NAND open source ftl, I would like to give a try to it, but if anyone already has some (bad) experiences with it let me know :)
18:32:25 Join markun [0] (~markun@rockbox/developer/markun)
18:32:51saratogai think that preset would be just one band, centered at 1khz, with a Q of about .2 or .3
18:37:42 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:37:57 Join Narod [0] (
18:46:48 Join ikeboy [0] (
18:47:07saratogathis is a single band with Q=0.2 @ 1KHz:
18:47:29 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
18:48:19pamaurycopper: how is the fuze+ doing ?
18:53:00coppersaratoga: yup, that's what I was intimating
18:53:25copperpamaury: about 24 hours in, 43% battery o_O
18:53:34copperI think it was at 46% *hours* ago
18:53:51pamaurymaybe the calibration is weird, what is the battery voltage ?
18:54:10copperis that in the debug menu somewhere?
18:55:21copper3.790 V
18:56:55pamaurywhat kind of file are you playing ? with headphones ? which volume ? does the whole album fits into the audio buffer ?
18:57:28 Quit ikeboy (Ping timeout: 245 seconds)
18:57:29copperexactly what we convened upon
18:58:28pamauryok, let's wait until the end, it will be quite interesting if you actually reach 40h !
18:58:58copperLAME V0, -20 dB, ear buds attached, the whole album is 303 MiB
18:59:32copperexact same conditions as the first battery benchmark on the Fuze+
19:00:09copper(and no replaygain / EQ / DSP running)
19:00:59copperalso, internal storage (I'm waiting for a replacement of my broken microsdxc card)
19:01:27copper16 GB model, dunno if batteries differ
19:01:43copperalso, refurb, so possibly not a brand new battery?
19:05:33 Quit lebellium (Quit: ChatZilla [Firefox 26.0/20131118212339])
19:18:06 Join kugel [0] (
19:18:06 Quit kugel (Changing host)
19:18:06 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:22:31 Quit kugel (Ping timeout: 246 seconds)
19:26:14 Quit markun (Ping timeout: 264 seconds)
19:27:53 Quit lorenzo92 (Ping timeout: 245 seconds)
19:38:32***Saving seen data "./dancer.seen"
19:48:02 Quit krabador (Quit: Sto andando via)
19:48:39 Join krabador [0] (
19:48:41 Quit krabador (Changing host)
19:48:41 Join krabador [0] (~krabador_@unaffiliated/krabador)
19:50:21 Join markun [0] (~markun@rockbox/developer/markun)
19:53:10copperpamaury: my Fuze+ is going stronger than the bloody Energizer bunny o_O
19:55:14pamaurycopper: :D
19:58:33copperpamaury: what is the supposed cause of extended battery life? That revision doesn't seem to feature touchpad disabling when locked
20:05:06pamaurycan you recall me the commit ?
20:05:19pamauryI think it has all the imx233 power management features on, nothing fancy
20:06:36 Join petur [0] (~petur@rockbox/developer/petur)
20:07:01 Quit shamus (Read error: Connection reset by peer)
20:07:24copperand what do you suspect is causing the battery life regression, for the Fuze+? What prompted you to run battery benchmarks?
20:07:33 Join shamus [0] (
20:08:21pamauryI'm not sure
20:08:49pamauryThe NWZ have very poor battery life, so I wanted to make sure if it was related to some power regression or it was nwz specific
20:09:02copperisn't it a bit much of a coincidence that both the Fuze+ and the Classic have regressed?
20:09:19coppersomeone should run a benchmark of another target
20:09:48copperI could run a battery benchmark for the Clip+, but I don't have a previous reference
20:09:48pamauryI don't know, on the fuze+ that's not that surprising given the number of changes in the past month
20:10:01pamauryI was excepting some regressions when adding support for the stmp3700 and stmp3600
20:10:34 Quit Strife89 (Quit: Heading out.)
20:11:08copperpamaury: assuming the regression spans different targets, what (older) commit do you think I should try for reference?
20:11:14coppercan you make an educated guess?
20:11:31pamauryNot out of the blue, no
20:11:44copperMeh. ;)
20:12:29NarodDo they, though? Would love to battery bench my two NWZs. :)
20:13:34pamaurybattery life on the nwz is very low at the moment: ~10hours
20:13:35copperto put it in other words: did anyone here commit something pertaining to CPU usage?
20:13:52copperor boosting or powersaving or whatever
20:14:20NarodWow, okay. I never really used it for that long in one go.
20:14:24copperanything that could remotely have an effect on battery usage
20:14:36pamauryI don't know, I try to follow on the commits but I don't remember anything special
20:14:43pamauryand it might be something less obvious
20:15:20copperok, this is what I'm going to do: I'm going to run battery benchmarks on the Clip+, with a year old revision, and the latest revision
20:15:35copperif I see a significant difference, then it's probably something target-agnostic
20:16:10copperI don't mind running battery benchmarks on the Clip+, because I never use it
20:17:09pamaurythe only thing which could be remotely be related is some change in the eq I think, but I think it's not enabled by default so it should not changed anything
20:18:52copperwell, if I don't see a significant difference with the Clip+, then at least I can rule out a target-agnostic commit
20:19:08copperI'll test the same revision as the iPod Classic
20:19:30pamauryon imx233 I see plenty of reasons why battery life could be worse than it used to be ^^
20:19:39copperit should help narrow it down, I think
20:20:01copperbecause the regression is definitely real
20:20:47copperI'm hoping the Clip+ benchmark will show the difference, because in that case, I can run as many benchmarks as I need, on the Clip+
20:21:21pamaurywhen my bench is done I will start two more benches on my two fuze+ for other revisisons
20:21:32pamaurycurrently the two revisions I chose were imx233 related
20:22:34copperI will use FLAC as the codec, in order to match my iPod Classic benchmarking conditions
20:22:47copperand because FLAC decoding is very fast
20:23:20copperI already expect a stupid bug to be the culprit
20:23:30copperunintended consequences
20:24:33 Quit rela (Ping timeout: 245 seconds)
20:29:37 Quit W0rmDr1nk (Ping timeout: 252 seconds)
20:48:16 Quit markun (Ping timeout: 272 seconds)
20:50:24 Join markun [0] (~markun@rockbox/developer/markun)
21:01:31 Quit markun (Ping timeout: 252 seconds)
21:02:53 Join dabgotra [0] (
21:03:06 Quit petur (Ping timeout: 272 seconds)
21:05:06 Nick DormantBrain is now known as SuperBrainAk (~andy@2001:470:8:a61::5f92:59a1)
21:07:49 Quit dabgotra (Quit: Leaving)
21:21:01 Join Tukeke [0] (~Tukeke@unaffiliated/hlvsv)
21:29:24*pamaury summons TheSeven
21:29:43 Join Jinx [0] (~Jinx@unaffiliated/jinx)
21:38:35***Saving seen data "./dancer.seen"
21:42:36 Quit krabador (Remote host closed the connection)
21:45:49 Join krabador [0] (~krabador_@unaffiliated/krabador)
21:56:14 Join wodz [0] (
21:59:40 Join markun [0] (~markun@rockbox/developer/markun)
22:00:22 Join Wardo [0] (
22:01:51 Join kugel [0] (
22:01:51 Quit kugel (Changing host)
22:01:51 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:02:47 Quit bluebrother (Disconnected by services)
22:02:52 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:16:29 Quit Wardo (Quit: Blarglarg)
22:18:22*TheSeven feels summoned
22:18:30*TheSeven summons pamaury
22:21:31 Join petur [0] (~petur@rockbox/developer/petur)
22:21:42 Quit petur (Client Quit)
22:21:56 Join petur [0] (~petur@rockbox/developer/petur)
22:22:11 Quit markun (Quit: leaving)
22:22:30 Quit petur (Client Quit)
22:24:14pamauryTheSeven: you know quite a bit of NAND flash right ?
22:27:37pamauryI'm wondering why some many codes dealing with NAND I have seen use tables about NAND chips instead of relying on the information provided by the chips themselves
22:28:18TheSevenI'd say mostly because the chips typically don't tell them all the parameters they need
22:28:25 Quit mudlord (Quit: mudlord)
22:29:09pamauryTheSeven: like what ? timing for example ?
22:29:18TheSeventhe driver tables typically contain much more information (size, geometry, timing contraints, parallel access capabilities, reliability information)
22:29:29wodzpamaury: AFAIK this is preONFI legacy stuff
22:31:25 Join benedikt93_ [0] (
22:34:23pamaurydo all modern chip follow ONFI spec ?
22:34:50wodzall I've seen datasheet for
22:35:17 Quit benedikt93 (Ping timeout: 272 seconds)
22:36:06 Quit kevku (Ping timeout: 260 seconds)
22:40:15wodzpamaury: Introducing ZLP fix for blk transfers kills usb in rb. Now I am testing only fix for ctr.
22:42:50pamaurywodz: I hope I don't say something stupid here but ZLP doesn't apply the same to bulk
22:43:06pamauryfor bulk transfers you send a ZLP only to prematuraly terminate a transfer
22:43:30pamauryif you are requested for 512 bytes, you must send 512 and no ZLP because that's the exact amount requested
22:43:36wodzpamaury: Hmm, it seems to work. I copied ~140MB file and spammed ep0 during copy with lsusb -vv in loop. The transfer survived and files are the same according to md5sum
22:44:02wodzI spammed ep0 during md5summing also
22:46:09pamaurywhat I don't understand is why it was failing
22:47:16pamaurybecause this zlp fix only apply to control transfers and most if not all transfers you send when spamming are not multiple of 64 in length, so no ZLP applies
22:47:22pamauryanyway, if it works, that's great :)
22:50:11 Nick benedikt93_ is now known as benedikt93 (
22:50:23 Quit benedikt93 (Changing host)
22:50:23 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
22:53:24wodzoh, I was able to kill the transfer with massive lsusb :-/
22:59:03wodzBUT it was much harder then before. I must check OF because I didn't try that hard to interrupt transfer before.
23:09:04 Quit amayer (Quit: Leaving)
23:17:15 Join fs-bluebot [0] (
23:24:37wodzpamaury: lsusb spam kills usb transfer in OF. I have a feeling that even easier then in rb
23:25:23wodzthe difference is that in rb we get usb reset while in OF we get usb disconnect
23:26:11pamaurywodz: this is getting interesting
23:28:06pamauryI wonder if it could be somehow related to back-to-back setup packets
23:28:43wodzyeah, rb driver with ZLP fix survives a few rounds of spamming while OF gets nuked by the first few spamming lsusbs
23:29:13[Saint]Hmmmm - clearing backdrops is definitely rather fucked up in ROckbox.
23:29:25[Saint]Not even changing themes works consistently.
23:29:36[Saint]Reboot after the fact does, though.
23:30:48wodzpamaury: spamming with lsusb while there is no transfer does not harm btw.
23:33:53pamauryI'm beginning to wonder if this is really a software bug...
23:34:53wodzme too
23:36:24wodzI'll post zlp thing to gerrit and ask mortalis to see if it helps
23:38:39***Saving seen data "./dancer.seen"
23:38:45 Quit ender` (Quit: Just as the absurd acts which prohibited the sale of game were virtually repealed by the poacher, just as many absurd revenue acts have been virtually repealed by the smuggler, so will this law be virtually repealed by piratical booksellers. -- Thomas B)
23:43:13 Join rela [0] (~x@pdpc/supporter/active/rela)
23:43:20 Quit rela (Read error: Connection reset by peer)
23:45:21 Quit Tukeke (Quit: Tukeke)

Previous day | Next day