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 2013-11-24

00:00:27n1swe pretty much need "put framebuffer on display"
00:00:39pamaurydecoding codecs on it would be pretty cool though
00:01:06[Saint]When they "opened" the display driver, they really just releaed an open shim to a closed source binary - which is still _something_ I guess, but yeah.
00:01:27[Saint]There's some painfully slow going on reversing the videocore binary.
00:01:55[Saint]Not going terribly fast, or far, though. Perhaps they need a pamaury. ;)
00:03:11 Quit ender` (Quit: Space, the last best hope for peace. These are the voyages of the Millenium Falcon. Its five-year mission: to find technology to defeat the Goa'uld.)
00:03:20[Saint]And while doing codecs on the GPU would be very cool - I think we're fast enough on ARM with most codecs now to not make that a major worry at all.
00:03:30[Saint]s/now/for a long time/
00:05:49 Quit onder` (Quit: leaving)
00:06:28pamauryis this videocore thing common or just on the rspi ?
00:06:40amiconnrpi also has the (in)famous videocore stuff like the ipod video?
00:06:46 Join onder` [0] (~onder@dyn-dsl-to-76-75-106-228.nexicom.net)
00:07:59pamauryamiconn: yeah, but a much older one
00:09:53*[Saint] is somewhat surprised that the raspberrypi hasn't restarted a new boom in the "Lyre" project...
00:10:30[Saint]You could sell kits. Raspi, PicoPSU, generic project box, bunch of buttons wired for a GPIO header, and a 7" composite LCD.
00:11:30[Saint]roll a tiny distro based on Arch or something suitable tiny. bam.
00:11:46pamaurybut it would be quite big
00:12:57[Saint]You could add a keyboard! :)
00:14:44 Quit lorenzo92 (Ping timeout: 248 seconds)
00:14:56wodzpamaury: hwstub for rk commited :-)
00:16:34pamaurygood
00:17:11wodznow the challenge is hwstub for atj :-)
00:19:36pamauryI have two: hwstub for rknano and tcc
00:20:26wodzpamaury: I was wondering how usb connected debuggers inform gdbserver about hitting breakpoint? I was thinking that we could patch mem to jump into hwstub but then it needs to somehow inform gdbserver that breakpoint is hit
00:22:05wodzpamaury: with rknano I would start from porting hwstub to amsv2
00:22:28wodzthen you will have some means of feedback from target during development
00:22:39pamauryusually you rather write an undefined instruction and capture this undef instr handler
00:23:06wodzpamaury: we could use swi as well. That part really doesn't matter
00:24:09pamauryso what is your question then ?
00:24:39wodzhow the gdbserver knows the breakpoint was hit
00:25:01pamaurypolling or we define an interrupt endpoint
00:25:28pamauryit would be preferrable to implement transport over bulk anyway, for speed reasons
00:26:24wodzCurrent implementation has BIG benefit of being very simple. Only control transfers.
00:26:34pamauryyes I know
00:26:49pamaurythat's why I wrote it this way
00:27:33pamaurybut to support remote debugging we would need to enhance hwstub quite a bit
00:27:43pamaurywe would need to intercept the irq handler and hide usb from the running code
00:30:55 Quit n1s (Quit: Ex-Chat)
00:37:27 Quit wodz (Quit: Leaving)
00:47:50shadows[Saint]: sadly, raspi nearest competitor (in my opinion) beaglebone black is minus an audio interface
00:49:05pamaurywhat are the differences ?
00:49:21pamauryraspi has jack and hdmi audio ?
00:51:24shadowsraspi has 3.5mm type audio jack; bbb has none and it also has audio output through hdmi
00:51:52[Saint]so does raspi
00:51:56[Saint](audio over hdmi)
00:51:59shadowsah, yes
00:52:47shadowsnot an expert myself, but bbb is also fully documented
00:52:51pamauryoh really, no jack ?
00:54:52shadowsjack exists on BegaleBone and BeagleBone XM ; those are also about $100 usd more costly
00:59:16 Nick SuperBrainAk is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
00:59:34 Quit mikroflops (Ping timeout: 245 seconds)
01:00
01:00:33 Join mikroflops [0] (~yogurt@s208h18o1esc1.dyn.tyfon.se)
01:07:15 Quit shamus (Read error: Connection reset by peer)
01:07:55 Join shamus [0] (~shmaus@ip-206-192-193-180.marylandheights.ip.cablemo.net)
01:21:29 Quit bertrik (Remote host closed the connection)
01:39:14***Saving seen data "./dancer.seen"
01:39:43 Join treaki_ [0] (bc31053766@p4FDF7E84.dip0.t-ipconnect.de)
01:42:12 Quit treaki__ (Ping timeout: 248 seconds)
01:51:33 Quit Tukeke (Quit: Tukeke)
02:00
02:04:33 Join Narod- [0] (Narod@p5DDDAE6A.dip0.t-ipconnect.de)
02:06:12 Quit Narod (Ping timeout: 246 seconds)
02:08:13 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
02:19:56 Quit Gallomimia (Quit: Gallomimia)
02:22:20 Quit Narod- ()
02:38:51pamaury[Saint]: I'm looking at the videocore project, they made a lot of progress recently
02:41:04[Saint]pamaury: #raspberrypi-internals is where most of the IRC stuffs happens afaik
02:42:05pamauryanyway I don't really plan to work on this, not in the near future
02:42:25pamauryor someone would have to find a very good reason for it
02:51:27 Join Gallomimia [0] (~gallomimi@216.232.233.122)
02:54:30 Join impossible [0] (~impossibl@user-1087s8g.cable.mindspring.com)
03:00
03:12:45 Quit Gallomimia (Quit: Gallomimia)
03:15:41 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
03:32:35shadowspamaury: very good financial reason? do you mean that?
03:39:16***Saving seen data "./dancer.seen"
03:53:22impossiblei installed rockbox on my fuze and I'm missing one song.
03:53:38impossibleit was in flac. Does rockbox play flac?
03:57:45impossibleWhoops. Turns out I deleted the file. Loving rockbox though
04:00
04:04:29 Quit cmhobbs (Ping timeout: 265 seconds)
04:30:11 Quit amiconn (Disconnected by services)
04:30:11 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:30:14 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:30:18 Quit pixelma (Disconnected by services)
04:30:19 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:30:21 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:32:42 Quit nosa-j (Ping timeout: 272 seconds)
04:34:03 Join nosa-j [0] (~m00k@66.233.224.206)
04:41:52 Nick DormantBrain is now known as SuperBrainAk (~andy@2001:470:8:a61::5f92:59a1)
04:42:53 Join Gallomimia [0] (~gallomimi@S0106000f6690b14a.ca.shawcable.net)
04:51:25 Quit Gallomimia (Ping timeout: 272 seconds)
05:00
05:00:15 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
05:03:25-->"oper #nsfw" received from thequux (~thequux@tx-67-77-76-133.dyn.embarqhsd.net)
05:03:45-->"help" received from thequux (~thequux@tx-67-77-76-133.dyn.embarqhsd.net)
05:04:35 Quit TheSeven (Disconnected by services)
05:04:50 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:06:09-->"test" received from thequux (~thequux@tx-67-77-76-133.dyn.embarqhsd.net)
05:18:26 Quit nosa-j (Ping timeout: 245 seconds)
05:22:07 Join nosa-j [0] (~m00k@66.233.224.206)
05:26:10 Quit cmhobbs (Ping timeout: 246 seconds)
05:39:08 Quit [Saint] (Remote host closed the connection)
05:39:20***Saving seen data "./dancer.seen"
05:40:14 Join [Saint] [0] (~saint@rockbox/user/saint)
05:43:37 Quit K1773R (Excess Flood)
05:43:57 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
06:00
06:03:17 Quit impossible (Remote host closed the connection)
06:10:02 Quit foolsh (Quit: foolsh)
06:39:13 Quit mc2739 (Quit: leaving)
06:42:56 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
07:00
07:00:56 Quit copper (Ping timeout: 252 seconds)
07:13:02 Quit thegeek_ (Ping timeout: 264 seconds)
07:33:10 Join mortalis [0] (~kvirc@213.33.220.118)
07:39:24***Saving seen data "./dancer.seen"
07:43:21 Join shai [0] (~Shai@l192-117-110-233.cable.actcom.net.il)
08:00
08:01:50 Quit Maxdamantus (Ping timeout: 260 seconds)
08:10:38 Join Maxdamantus [0] (~Maxdamant@2001:470:f078::dead:beef:cafe)
08:11:33 Quit shai (Quit: Leaving)
08:15:52 Join copper [0] (~copper@unaffiliated/copper)
08:17:14 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
08:26:56copperbattery benchmark with a recent revision on the Clip+, with lossyFLAC: https://outpost.fr/tmp/d6V.txt
08:46:38 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:51:20[Saint]Whoah.
08:51:35[Saint]SO much different looking at a target where the estimated runtime is even vaguely correct.
08:53:29copper[Saint]: is there any chance that the reason is that the Clip+ is never "unboosted"?
08:53:43copper(if I recall correctly?)
08:54:30[Saint]...huh?
08:54:44[Saint]I'm not sure how those would relate to each other, even if it were true.
08:55:47copperjust some bit that I remember from the test_codec benchmarks that saratoga asked me to run some time ago
08:56:15copperdunno if the "boosting / unboosting" thing only concerned the test_codec plugin
08:56:38[Saint]I think you may be misremembering. The CLip+ certainly is "never unboosted", it very rarely boosts. And this doesn;t have anything to do with estimated runtime.
08:57:01*copper greps his logs
08:57:14[Saint]Errr, dammit. I worded that very oddly. The Clip+ has very little reason to ever boost.
08:57:37[Saint]Except if you're running some truly insane codec and to fill the buffer.,
08:58:30[Saint]You can run run an unboosted test_codec run, and that would result in awful runtime...perhaps that's what you're thinking of?
08:59:02copper<saratoga> yeah the clip+ has boosting disabled
08:59:24copper2013/07/28 20:28:43 UTC
08:59:47copperbut that related to the test_codec plugin
08:59:52copper<copper> I don't see that option on the Clip+
09:00
09:00:12copper<saratoga> it was a bit unstable and we never figured out why
09:02:44copper[Saint]: I'm just talking out of my ass, like most of the time ;)
09:02:59[Saint]Hmmm..I see. I don't remember that sticking. I thought that was only temporary.
09:03:07coppertrying to related bits and pieces I know nothing about and don't understand :-P
09:03:17copperrelate*
09:03:58copperanyway… my Clip+ is charging, and then I'll run the year old build like the Classic
09:04:14coppersince the Clip+ battery is so much smaller, I'll be able to get the results today
09:04:21[Saint]I think I recall the Clip+ not getting terribly much out of running an unboosted clock anyway. But now I'm struggling to recall because I thought it still had boosting enabled.
09:05:53[Saint]Well, my apologies. I really thought that was only temporary, I certainly recall it having boosting once upon a time.
09:08:36[Saint]Maybe I'm mixing things I think I recall about the CLip+ and the ClipV1
09:12:06[Saint]So much variance in the SansaRuntime Clip+ stats its pretty much impossible to gain anything useful from it.
09:12:55 Join thegeek [0] (~thegeek@60.36.34.95.customer.cdi.no)
09:13:17[Saint]9h in 2010, to 17h in 2010, 14h in 2011, to 10h in 2013, and now your 15h in 2013. :-/
09:14:34copperI guess you can only reliably compare benchmarks on the same unit, at about the same time
09:14:45copperlike what I'm doing now
09:15:20[Saint]I'll dig up and charge a Nano2G this evening when the So goes to bed.
09:15:24coppersame unit at the same time = same battery at the same age
09:15:35[Saint]Since tests there will take changing builds and looking at a screen.
09:15:42[Saint]Rather than waiting ~20h
09:15:51[Saint]+
09:15:53copperwhat is the So?
09:16:10[Saint]Significant Other - Ms [Saint] :)
09:16:14copperah
09:16:26copperand what's special about the Nano 2G?
09:16:49[Saint]Its the only target that can display its exact power usage.
09:16:57coppernice
09:16:58[Saint](to my knowledge)
09:18:08[Saint]There's about what, ~300 commits to cover in the period you're testing from?
09:18:27[Saint]git bisect should have that done in no time.
09:19:12[Saint]for reference for me later on, what is the closest "good" commit?
09:19:53copperiPod Classic: 3e1c492
09:20:07copperFuze+: c4f2a46
09:20:20copperfirst is from november 2012, second is from July 2013
09:20:45copperbut right now I have no idea if there is any relation between the two in terms of battery life
09:20:45[Saint]So its variable by device, or those are just builds you happened to have laying around?
09:21:11copperI chose the Classic commit from 2012 because that's what I tested a year ago
09:21:37copperpamaury chose the Fuze+ commit because he knew that one was supposed to have long battery life, but I think that was Fuze+ specific
09:22:16[Saint]Ok, nevermind. I'll pick an arbitrary point in time that predates both. Easier.
09:22:25copperI'll be testing the Clip+ with the same year old commit as the Classic, to see if the battery life difference is target-agnostic or not
09:23:08[Saint]I would find it hard, but not impossible, to believe that a bunch of device-specific regressions were introduced.
09:23:20copperand yeah, it's a pain to test, but that year old commit was my only known reference
09:24:01[Saint]Looking over the commits from that period, there's no immediate show-stopper that jumps out and says "test me".
09:24:07[Saint]But, there;s a few I could try.
09:24:16[Saint]I'll let git biscet descide.
09:24:37copperif it was more or less obvious, I suspect someone else would have found out about it already ;)
09:25:29copperlike I told GodEater the other day, a 25% drop in battery life is not necessarily obvious when you're casually using your DAP
09:25:32[Saint]You'd think. But stranger things can happen.
09:25:48[Saint]And, no. It isn't.
09:26:12[Saint]I charge all the time, whenever I can.
09:26:28[Saint]So I wouldn't ever notice it unless battery time dipped below ~5 hours.
09:27:16[Saint]Hmmmm...there _was_ a really big playback engine reshuffle....
09:28:33copper[Saint]: btw, the battery life difference with the Fuze+ is rather interesting:
09:28:37copperhttps://outpost.fr/tmp/laE.txt
09:28:40copperhttps://outpost.fr/tmp/LUN.txt
09:29:01copper41 hours in july 2013 vs 30 hours now
09:29:25copperso if the problem is target-agnostic, there may be a chance that it came after july 2013
09:30:06copperthough the Fuze+ is a slightly different beast, since there has been a lot of work semi recently to improve its battery life
09:30:17copperbut the drop is large enough to be worth mentionning
09:30:50coppera Fuze+ build from late 2012 would also likely run for 30 hours, because the improvements came later
09:31:15copperthe more devices we collectively test, the better, I guess
09:39:25***Saving seen data "./dancer.seen"
09:40:07[Saint]After July?
09:40:15[Saint]Well..that's my candidate excluded than.
09:40:22[Saint]*then
09:40:23 Join Gallomimia [0] (~gallomimi@99.199.8.77)
09:42:31[Saint] g#479 ?
09:42:33fs-bluebotGerrit review #479 at http://gerrit.rockbox.org/r/479 : by Michael Sevakis (changes/79/479/9)
09:44:34coppermy config sets 44.1 kHz
09:46:12[Saint]If I assume target has nothing to do with it, and take your "July-ish" feeling into account...that seems like the likliest change to accidentally balls something up - though it shouldn't, to my knowledge.
09:49:01[Saint]I initially suspected the playback engine rework, but with the data you've given that is ruled out.
09:49:28copperthat means running two more benchmarks on any given target: one with that revision, and one with the revision before that one
09:50:21[Saint]I'm not going to go with any gut hunches on this one, I'll end up leading myself astray.
09:50:56[Saint]I'll pick an arbitrary point in time in the "known good" region and let observation and git bisect decide.
09:51:41[Saint]If I don't manage to sneak away tonight I should have all day tomorrow.
09:52:10copperhmmm, but that commit predates the 40 hours Fuze+ revision
09:53:03[Saint]When it was committed to gerrit, yes.
09:53:07[Saint]To Rockbox, no.
09:53:57[Saint]It was committed between the 2nd and 5th of July 2013
09:54:09coppergit log puts it online 3112, while the Fuze+ revision is on line 2536?
09:54:27copperthe Fuze+ revision was committed on July 23
09:54:54[Saint]Ahhh, well. Shit.
09:55:25copperunless that IS indeed the offending commit, and instead of 40 hours, the Fuze+ should have gone 50 hours!
09:55:38[Saint]Hahahaha :)
09:55:48[Saint]...possible.
09:56:20 Quit soap (Ping timeout: 272 seconds)
09:56:26*[Saint] now ses where he misread a datestamp
09:56:32 Quit kevku (Read error: Connection reset by peer)
09:56:33[Saint]*sees
09:56:47 Join ender` [0] (krneki@foo.eternallybored.org)
09:56:58[Saint]Serve's me right for going looking again for possibly-suspect commits.
09:57:41 Quit mortalis (Read error: Operation timed out)
09:58:42[Saint]Knowing my luck I'll test on N2G when I get a chance and everything will be kosher.
09:59:12 Quit ender` (Client Quit)
09:59:38copperugh, that is a big commit
10:00
10:01:38[Saint]But if one assumes your F+ revision is "good" it can't be the offender.
10:02:29[Saint]Disregarding that info, it certainly looks like a suspect. It walks over enough things.
10:03:14 Join lorenzo92 [0] (~chatzilla@host214-107-dynamic.40-79-r.retail.telecomitalia.it)
10:03:16[Saint]I'll take off the speculation hat until I can do something about it. Thinking about possibilities is driving me nuts.
10:03:25copper[Saint]: the Fuze+ revision is special because it includes work on saving battery; there's no telling if the improvements were so large that they would absorb the regression.
10:04:41copperIf the clip+ with the year old revision shows better battery life, I'll try the sampling rate commit on it right after
10:05:42copperI'd rather test the Clip+ over my Fuze+ and Classic because 1) I don't use it, and 2) its battery is smaller and thus the benchmark last half as long
10:05:47copperbenchmarks*
10:07:21 Join soap [0] (~soap@rockbox/staff/soap)
10:09:12 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
10:10:28 Quit lorenzo92 (Ping timeout: 248 seconds)
10:10:43 Join ender` [0] (krneki@foo.eternallybored.org)
10:12:12copperer
10:12:25copperthat wouldn't explain the Fuze+ regression
10:12:47copperif it lasted 40 hours WITH the regression, then its later regression would not be related
10:12:59coppersigh
10:13:26 Quit soap (Ping timeout: 272 seconds)
10:19:33 Join lorenzo92 [0] (~chatzilla@host214-107-dynamic.40-79-r.retail.telecomitalia.it)
10:43:48 Join markun [0] (~markun@rockbox/developer/markun)
10:49:30 Join Wardo [0] (~Mirandaha@176-120-190-109.dsl.ovh.fr)
10:54:26 Join bertrik [0] (~quassel@ip117-49-211-87.adsl2.static.versatel.nl)
10:54:26 Quit bertrik (Changing host)
10:54:26 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:00
11:04:46 Join Narod [0] (Narod@p5DDDAE6A.dip0.t-ipconnect.de)
11:12:17 Nick SuperBrainAk is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
11:19:07 Quit Wardo (Quit: Blarglarg)
11:39:28***Saving seen data "./dancer.seen"
12:00
12:09:20 Quit lorenzo92 (Quit: ChatZilla 0.9.90.1 [Firefox 25.0.1/20131112160018])
12:59:45pamaurycopper: how is the clip+ doing ? finish ?
13:00
13:00:12pamauryOne on my fuze+ finished the after 7h ! That the fuze+ I only rarely use, maybe the battery is dead :(
13:00:17copper07:26:57 UTC <copper> battery benchmark with a recent revision on the Clip+, with lossyFLAC: https://outpost.fr/tmp/d6V.txt
13:00:35copper15 hours
13:01:19 Quit K1773R (Read error: Operation timed out)
13:03:49pamauryis that good or bad ?
13:04:25copperthat's fairly normal I think
13:04:42copperexcept I'm using (lossy) FLAC, which is very efficient
13:05:01copperright now I'm running another benchmark on the Clip+ with the same year old revision that I ran on my Classic
13:05:08copperwe'll see if there is a significant difference
13:08:18copperfor reference, FLAC on the Clip+ decodes nearly 4 times faster than LAME
13:08:33copperand my lossyFLACs are barely twice the size
13:09:31 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
13:28:01 Join lorenzo92 [0] (~chatzilla@host214-107-dynamic.40-79-r.retail.telecomitalia.it)
13:39:30***Saving seen data "./dancer.seen"
14:00
14:12:21 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:30:49 Quit kugel (Ping timeout: 252 seconds)
14:34:04 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
14:38:54 Join krabador [0] (~krabador_@host235-196-dynamic.181-80-r.retail.telecomitalia.it)
14:38:54 Quit krabador (Client Quit)
14:39:15 Quit lorenzo92 (Ping timeout: 248 seconds)
14:51:39 Quit mc2739 (Quit: leaving)
15:00
15:01:31 Join soap [0] (~soap@rockbox/staff/soap)
15:05:25 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
15:05:48 Quit mc2739 (Remote host closed the connection)
15:13:28 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
15:23:00 Join rela [0] (~x@pdpc/supporter/active/rela)
15:24:24 Join krabador [0] (~krabador_@unaffiliated/krabador)
15:39:33***Saving seen data "./dancer.seen"
15:49:59 Quit soap (Ping timeout: 272 seconds)
15:52:56 Join soap [0] (~soap@rockbox/staff/soap)
15:54:47 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
16:00
16:00:23 Quit soap (Ping timeout: 245 seconds)
16:03:21 Join soap [0] (~soap@rockbox/staff/soap)
16:29:06 Quit rela (Read error: Connection reset by peer)
16:37:39 Quit cmhobbs (Ping timeout: 248 seconds)
16:40:58 Join cmhobbs [0] (~cmhobbs@ip70-178-52-92.ks.ks.cox.net)
16:40:58 Quit cmhobbs (Changing host)
16:40:58 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
16:45:45 Quit cmhobbs (Ping timeout: 252 seconds)
17:00
17:27:56 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
17:39:34***Saving seen data "./dancer.seen"
17:47:36 Join ikeboy [0] (~ikeboy@ool-435622d3.dyn.optonline.net)
18:00
18:08:55 Join Guest81512 [0] (Slayer@c-69-143-178-62.hsd1.va.comcast.net)
18:12:35 Quit Guest11127 (Ping timeout: 272 seconds)
18:39:04 Quit krabador (Quit: Sto andando via)
18:46:47 Join linuxguy3 [0] (~timj@24-148-61-208.c3-0.lem-ubr1.chi-lem.il.cable.rcn.com)
18:48:43 Quit ikeboy (Ping timeout: 245 seconds)
19:00
19:01:06 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:05:48 Quit cmhobbs (Ping timeout: 245 seconds)
19:10:18 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
19:34:52 Quit cmhobbs (Ping timeout: 265 seconds)
19:39:39***Saving seen data "./dancer.seen"
19:42:11 Quit markun (Ping timeout: 248 seconds)
19:42:23 Join markun [0] (~markun@rockbox/developer/markun)
19:43:27 Join cmhobbs [0] (~cmhobbs@ip70-178-52-92.ks.ks.cox.net)
19:43:29 Quit cmhobbs (Changing host)
19:43:29 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
19:43:46 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
19:53:23 Quit cmhobbs (Ping timeout: 248 seconds)
20:00
20:08:23 Quit kugel (Ping timeout: 246 seconds)
20:18:05copperpamaury: clip+ currently at 22%, but that's expected: I started the benchmark somewhat late
20:18:26copperer, I'm replying to the same question for the second time
20:18:31copperbacklog
20:18:32 Join Strife89 [0] (~Strife89@adsl-98-80-218-160.mcn.bellsouth.net)
20:19:28 Nick DormantBrain is now known as SuperBrainAk (~andy@2001:470:8:a61::5f92:59a1)
20:21:17pamaurycopper: on my fuze+ I think I will reach 30h but no more
20:26:35 Join Cultist [0] (~CultOfThe@c-24-12-53-28.hsd1.il.comcast.net)
20:26:36 Join saratoga [0] (123e1c08@gateway/web/freenode/ip.18.62.28.8)
20:26:43CultistHello
20:27:29CultistI wondered if anyone could recommend a music player with good rockbox support, something 16 or 32gb with good build quality in the $100~ range
20:27:31saratogaif you want to test battery life a lot, i'd recommend a DMM and a USB cable
20:27:39saratogatotal cost on Amazon.com is less than 10 USD
20:28:43saratogaCultist: probably a sandisk player and a microsd card
20:28:55copperpamaury: is your current Fuze+ revision older or newer than the 40 hours one?
20:29:18CultistI've got a Sandisk zipclip with rockbox on it, but it keeps freezing up and crashing when I try to play ogg files
20:30:14saratogadoes it always happen with the same files?
20:30:16n1sCultist: any ogg files or sepcific ones?
20:30:21saratogai'm surprised, ogg support is quite stable
20:31:01Cultisthappens with all the ones I've tried
20:31:24Cultistall converted from flacs, I guess it's possible something was going wrong with the conversion
20:31:27Cultistbut they play fine on mycomputer
20:31:36saratogaat the same place in each file?
20:31:42saratogaor is it random
20:31:42Cultistbefore they start
20:31:47CultistI open the file, everything freezes
20:31:50saratogaah thats probably bad files than
20:32:07Cultistexcept they work on my computer's media player
20:32:13n1swe still shouldn't freeze
20:32:21saratogatry this file and see if it works: http://download.rockbox.org/test_files/vorbis_128.ogg
20:32:54Cultistgotta dig up a microsd card, misplaced the one I had in there (I haven't messed with it in a while)
20:35:20pamaurycopper: all revisions we are testing are older than the 40hours one
20:42:39Cultisthm
20:42:42Cultistit works now
20:42:55CultistCould the filenames cause a problem?
20:43:08Cultistthings with overly long names, or with accented characters?
20:44:07saratogano those won't matter
20:44:34saratogaso that test file works, but your files do not?
20:45:54saratogacheap DMM: http://www.amazon.com/gp/product/B005EK3NRS/ref=s9_simh_gw_p60_d0_i1
20:46:09gevaertsLarge vorbis comment stuff, maybe?
20:46:15bertrikor perhaps weird/overly large metadata
20:47:07n1sthe problem with huge metadata is supposed to be fixed
20:47:52n1sanyway i'd be interested in a file that freezes the player with a recent version of rockbox
20:48:04saratogawow someone actually makes a USB power meter: http://www.amazon.com/PortaPow-Monitor-Multimeter-Ammeter-Chargers/dp/B00DF2485S/ref=sr_1_1?s=hi&ie=UTF8&qid=1385322439&sr=1-1&keywords=usb+power+meter
20:48:12saratogawonder if its any good
20:49:02n1s"now accurate to 0.01A" doesn't sound super useful
20:49:31 Quit y4n (Quit: Today is the perfect day for a perfect day.)
20:49:45saratogaah yeah too low to be that useful
20:49:51saratogaerr, our currents are too low
20:50:49saratogaa real DMM is cheaper and more versatile anyway
20:51:56 Join onder`_ [0] (~onder@dyn-dsl-to-76-75-119-170.nexicom.net)
20:55:21 Quit onder` (Ping timeout: 272 seconds)
20:55:27 Nick onder`_ is now known as onder` (~onder@dyn-dsl-to-76-75-119-170.nexicom.net)
21:00
21:01:10Cultistwent afk
21:01:23Cultistsaratoga, yeah. Or at least, they used to not
21:01:36saratogaso they work now?
21:01:36CultistI need to go get a new higher capacity microusb card, biggest one I can find at home is 8gb
21:01:46CultistI must have misplaced my old one with my music
21:02:04CultistI'll rebuild the set on a new card and see if they work now
21:03:29 Join Szczepancio [0] (~Patryk@acfl132.neoplus.adsl.tpnet.pl)
21:04:09Szczepanciopamaury: Did you make anything on linux based sony players?
21:05:59pamaurynot yet
21:06:02Szczepanciopamuary: I mean, did you anything after my absence.
21:06:35SzczepancioMeh, bad.
21:07:18SzczepancioI know it's not your fault, but it's bad that nobody take interest in sony linux mp4 player's. They are great.
21:08:58pamauryit's not that nobody is interested, it is that nobody has time
21:09:54saratogayeah, there are a lot of mp3 players out there
21:23:14 Join kugel [0] (~kugel@46-127-48-168.dynamic.hispeed.ch)
21:23:14 Quit kugel (Changing host)
21:23:14 Join kugel [0] (~kugel@rockbox/developer/kugel)
21:27:50 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
21:29:31 Quit saratoga (Quit: Page closed)
21:35:16 Join wodz [0] (~wodz@87-207-223-0.dynamic.chello.pl)
21:36:09wodzpamaury: I have problems compiling qeditor http://pastie.org/8505799
21:37:12pamaurywodz: give me 5/10 minute and I'll be available
21:39:43***Saving seen data "./dancer.seen"
21:41:39 Quit Szczepancio (Quit: Leaving)
21:50:13pamaurywodz: it compiles fine here :-/
21:51:19pamauryah I see, I'm using a gcc extension syntax
21:51:58wodzqmake emitted warning also: "Project MESSAGE: Warning: unknown QT: widgets"
21:52:29pamaurywodz: change the offending lines to this: http://paste.pm/c59.c
21:55:15wodzpamaury: works, thanks
21:59:52 Join SYSTEM_ [0] (~chatzilla@adsl-99-120-26-33.dsl.dytnoh.sbcglobal.net)
22:00
22:00:00 Part SYSTEM_
22:00:14 Join Water [0] (~chatzilla@adsl-99-120-26-33.dsl.dytnoh.sbcglobal.net)
22:02:14WaterHey guys. Anyone in here use a Fuze+?
22:03:19pamauryplenty
22:03:51WaterIs this a known bug: turn on Fuze+ (start screen set to "Resume Playback"), as soon as playback starts, long-press the "Play/Pause" button. RB now hangs for 20 seconds and nothing can be done.
22:03:54 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
22:03:57 Quit bluebrother (Disconnected by services)
22:04:26WaterI'm guessing it has something to do with the bookmark file.
22:04:46pamaurythat rings a bell but I'm note sure
22:05:01pamauryI'll try it when my fuze+ has finished my battery bench
22:05:17WaterCool. Ty.
22:05:23 Quit fs-bluebot (Ping timeout: 245 seconds)
22:06:42wodzpamaury: qeditor doesn't work for me. With my xml file it spits "parser error : Extra content at the end of the document </xml>" loading stmp files crashes the whole thing "GLib (gthread-posix.c): Unexpected error from C library during 'pthread_setspecific': Invalid argument. Aborting."
22:06:44 Join fs-bluebot [0] (~fs-bluebo@g225254033.adsl.alicedsl.de)
22:06:59pamauryhum
22:07:01pamauryweird
22:09:05wodzpamaury: my description file: http://pastie.org/8505850
22:13:27pamaurygoogle chrome doesn't like your file either
22:14:18pamauryah, you don't need to close the <xml> tag
22:14:46pamauryactually you must not close it
22:16:10wodzpamaury: that is contrary to what your XML.txt file states :-)
22:17:12wodzpamaury: after removing </xml> from my file it crashes qeditor the same as your stmp xml description files
22:18:02pamaurycan you run valgrind ?
22:18:20pamauryor gdb
22:18:40pamauryby the way, if you put your xml file into utils/regtools/desc/
22:18:47pamauryit will automatically be loaded
22:18:57 Join cmhobbs [0] (~cmhobbs@ip70-178-52-92.ks.ks.cox.net)
22:18:58 Quit cmhobbs (Changing host)
22:18:58 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
22:19:46pamauryit should be named regs-rk27xx.xml probably, for this to work
22:20:21pamauryI have no idea why it crashes though
22:20:24wodzbacktrace of the crash: http://pastie.org/8505870
22:20:56wodznot very helpful I guess
22:23:30 Quit Strife89 (Ping timeout: 240 seconds)
22:25:12pamauryindeed
22:25:16pamaurymaybe ask bluebrother^
22:25:22pamauryhe knows a lot about QT
22:30:02wodzvalgrind summary: http://pastie.org/8505890
22:33:50pamauryso the program is not doing anyway strange
22:33:56pamauryI mean appart from crashing ^^
22:35:04wodzyeah, without that minor detail it works great :P
22:35:14wodz*besides
22:35:42pamauryI never pretended that it actually works !
22:35:57pamauryI just pretended that it does things on my computer :p
22:36:04 Join Strife89 [0] (~Strife89@adsl-98-80-219-190.mcn.bellsouth.net)
22:40:10pamaurywodz: maybe it's a problem with QT ?
22:40:24pamauryor maybe it's related to this warning in qmake
22:40:49 Quit benedikt93 (Quit: Bye ;))
22:40:58wodzpamaury: I don't know C++ and don't know QT so can't say
22:41:18wodzpamaury: which version of qt do you have on your machine?
22:41:30pamauryi'm not QT expert some maybe I did something wrong
22:41:32pamauryQT4
22:42:22pamaury4.8.5 iirc
22:44:05wodz4.8.1 here
22:56:35 Quit markun (Ping timeout: 272 seconds)
23:00
23:06:13pamaurycopper: fuze+ ran for 26h :-/
23:07:21 Quit kugel (Ping timeout: 272 seconds)
23:10:11 Quit n1s (Quit: Ex-Chat)
23:11:11 Quit Gallomimia (Quit: Gallomimia)
23:13:16Waterpamaury: What file type/bit-rate? I got 33 hours 6 months ago when mine was brand new.
23:13:40Water~224 kb/s mp3
23:14:05pamauryWater: we are bisecting a power regression, this test does not represent the current status
23:14:15Waterah
23:14:19pamaurythe best we have obtained is ~40h with high quality mp3
23:14:36Waterwow cool!
23:15:30pamaurybut we have regressed since then and we trying to find why
23:15:49WaterCould I help with my Fuze+?
23:16:44Waterpamaury:
23:17:17pamauryI'm waiting for the results of copper but yeah definitely
23:17:48pamaurybest you can do is pop up tomorrow, I'll send you the files for a specific revision and explain you the procedure
23:17:53pamauryif you don't mind :)
23:17:59WaterGlad to help.
23:18:16WaterAround what time tomorrow?
23:19:02WaterSend the links.
23:19:06pamauryany time, I'll always connected during the day (GMT time)
23:19:10pamaury*I'm
23:20:56WaterOops, i misunderstood. I'll pop in tomorrow ready to go.
23:21:51pamauryok thanks
23:22:00 Join foolsh [0] (~foolsh@c-24-14-134-34.hsd1.in.comcast.net)
23:23:21 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:24:14WaterI have a vague memory of some commit comment months ago with words something like "remove all frequency scaling". I didn't actually read the code. We do scale CPU frequency on the Fuze+ though, right?
23:25:21 Quit lebellium (Ping timeout: 272 seconds)
23:25:32 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:27:44 Join pubwsli [0] (44bf2bb6@gateway/web/freenode/ip.68.191.43.182)
23:28:18 Quit pamaury (Ping timeout: 245 seconds)
23:32:28 Quit bertrik (Remote host closed the connection)
23:35:00pubwsliHello, I'm experimenting with addons in rockdoom and I keep encountering the same error. While Doom runs fine, my addon does not. I downloaded the .wad I wanted and put it in .rockbox/doom/addons.
23:35:18pubwsliI loaded Rockdoom, selected my addon, and pressed play game. Every time I do this, the lines of text flash as usual, but it says in one of them that something is wrong with my shareware and then it gets stuck on a line about sprites, duplicating it every second until it's been about 30 seconds, then rockdoom quits.
23:35:45pubwsliDoes anyone know what could be the issue?
23:38:37 Quit cmhobbs (Ping timeout: 252 seconds)
23:39:46***Saving seen data "./dancer.seen"
23:43:33 Join JdGordon [0] (cb1380e2@rockbox/developer/JdGordon)
23:43:43foolshWater: Yes the Fuze+ does frequency scaling. You probably remember this (Tue Sep 4 00:35:58 2012 "imx233: disable cpu frequency scaling") it was disabled but on (Fri Jan 18 18:59:08 2013 "imx233: enable cpu frequency scaling on all targets") it was enabled
23:44:21pubwsliI figured I had a bad .wad so I found another, but I got the same result. Where did I go wrong?
23:47:05 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
23:56:05 Join kugel [0] (~kugel@rockbox/developer/kugel)

Previous day | Next day