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 2012-10-18

00:01:29***Saving seen data "./dancer.seen"
00:01:34 Join [Saint] [0] (~quassel@rockbox/user/saint)
00:02:47 Quit XavierGr (Read error: Connection reset by peer)
00:04:58 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
00:09:39CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:09:39*the-kyle uses wavpack 44.1KHz mono when recording with mic input. It doesn't run my battery as much as stereo 48KHz would. It also probably doesn't produce as much noise from writing to flash as wav.
00:11:10the-kyleIn addition, I still benefit from the lossless recording, so that I can reencode to any other supported format using an encoder on my computer without losing audio quality.
00:19:30 Quit pamaury (Ping timeout: 260 seconds)
00:19:58 Quit Clear_runway (Ping timeout: 252 seconds)
00:22:33 Join JdGord [0] (~AndChat80@
00:28:12 Quit shamus (Read error: Connection reset by peer)
00:28:45 Join JdG [0] (~AndChat80@
00:28:53 Quit JdGord (Ping timeout: 260 seconds)
00:31:12 Quit JdG (Read error: Connection reset by peer)
00:31:16 Quit XavierGr (Read error: No route to host)
00:32:41 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
00:33:16 Join JdGord [0] (~AndChat80@
00:34:25 Quit JdGord (Client Quit)
00:34:50 Quit ender` (Quit: If man could be crossed with the cat it would improve man, but it would deteriorate the cat. -- Mark Twain)
00:35:54 Join JdGord [0] (~AndChat80@
00:40:06 Quit sentriz (Quit: Leaving)
00:42:27 Quit bertrik (Ping timeout: 255 seconds)
00:43:09 Join JdG [0] (
00:46:30 Quit JdGord (Ping timeout: 255 seconds)
00:49:24 Join Scr0mple [0] (~Simon@
00:49:24 Quit Scr0mple (Client Quit)
00:54:20 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
00:56:14speckmade1the-kyle: well, I'll surely try again with WavPack at 44.1 k sampling rate.
00:56:58speckmade1Last time I didn't care about quality that much - just about intelligibility.
00:57:40speckmade1and about being sure to get enough recording time.
00:58:02speckmade1what noise are you talking 'bout?
00:58:33speckmade1real audible noise in the recording? or was that figuratively speaking?
01:01:07the-kyleThere is a constant audible buzz when the backlight is on, and some buzzing/scratching noize when writing to flash. It's not unbearable, but it is audible. It's a hardware thing; I believe the OF does this also.
01:01:52speckmade1with 8 k sampling rate I had a hard time to even figure out the meaning of my own speech - with 44.1 k it would have surely been possible to get most words from people speaking in the background - that would surely be a feature!.. :-)
01:03:52the-kyleYes. On the clip+, the lower the recording frequency, the more ringing is present in the recording. It becomes noticeable below 44.1KHz and makes the recording unintelegible at 8KHz. Again, I believe this is hardware, and there's little rb can do to fix it. I just there were other lossy encoders other than mp3.
01:04:09the-kyles/just/just wish
01:04:35speckmade1WavPack in standard setting should be much lighter on the battery than LAME, I guess.
01:04:44speckmade1(Is it LAME?..)
01:05:24AlexPPretty sure Rockbox is libmad not lame for mp3 encoding
01:05:27speckmade1Isn't there? Why? I was wondering if Vorbis would be a more logical choice...
01:06:13AlexPmaybe that's playback thinking about it
01:06:20the-kyleVorbis would be better, but rb doesn't have an vorbis encoder in the recording settings. May be too expensive in ram or cpu to implement, but not sure.
01:06:45speckmade1I thought the "d" in "mad" was for "decoder"... :-)
01:06:51AlexPShine for encoding maybe
01:06:54*the-kyle I thought the mp3 encoder was lame, but I'm not sure about that either, as I generally avoid mp3.
01:07:00AlexPEither way, not lame I don't think
01:07:19AlexPAlthough I'm probably wrong :)
01:09:01derfthe-kyle: Perhaps Opus?
01:09:28AlexPthe-kyle: I think back in the day vorbis didn't have an integer encoder implementation
01:09:35AlexPNo idea if that changed
01:09:35derfIt still doesn't.
01:09:39derfAt least not a public one.
01:09:50AlexPright, that'll be why then :)
01:09:57derfOpus does.
01:10:40 Quit ender| (Ping timeout: 244 seconds)
01:11:57 Quit JdG (Quit: Bye)
01:13:56speckmade1that'd surely be the next unimaginable level of awesomeness... :-O
01:16:45 Quit AlexP (Ping timeout: 246 seconds)
01:17:38 Join AlexP [0] (~alex@rockbox/staff/AlexP)
01:21:14 Join zoktar [0] (
01:22:51 Join ender| [0] (~ender1@2a01:260:4094:1:42:42:42:42)
01:25:42 Quit lebellium (Quit: ChatZilla 0.9.89 [Firefox 17.0/20121010150351])
01:35:38 Quit ender| (Ping timeout: 246 seconds)
01:49:12 Join ender| [0] (~ender1@2a01:260:4094:1:42:42:42:42)
02:01:32***Saving seen data "./dancer.seen"
02:23:35 Join shamus [0] (
02:28:24 Join eckoit [0] (
02:29:53 Quit ender| (Ping timeout: 246 seconds)
02:33:50 Join saratoga [0] (~AndChat19@2600:1000:b006:5c02:0:2d:f615:d301)
02:34:12saratogaRockbox uses shine for mp3 encoding iirc
02:34:40saratogaSince we have floating point we need integer encoders
02:38:24saratogaDon't have floating point
02:41:49derfsaratoga: Are you guys planning to look into Opus encoding at some point?
02:42:05derfI forget if we've already talked about this.
02:42:25saratogaIt'd be welcome but I'm not too interested in it
02:42:54saratogaI don't do much with recording
02:43:06saratogaAlthough opus seems logical
02:43:17 Join ender| [0] (~ender1@2a01:260:4094:1:42:42:42:42)
02:52:06 Nick Raptors_ is now known as Raptors (~Raptors@
03:02:37 Quit saratoga (Quit: Bye)
03:18:43 Join Scr0mple [0] (~Simon@
03:21:37 Quit Scromple (Ping timeout: 256 seconds)
03:30:22 Quit eckoit (Quit: eckoit)
03:39:10 Join Clear_runway [0] (~Clear_run@
03:59:18*the-kyle would definitely use Opus if it was available. I'm probably not the person who would be able to submit the encoder for review, but I can certainly patch it in, test and report any bugs/crashes/less than awesome behavior I can find.
03:59:56[Saint]I'm fairly certain it wouldn't hit HEAD at present even if you did.
04:00:13[Saint]AFAIK, the consensus was to implement decoding in full first.
04:01:11[Saint]I think it's fairly safe to say that (if targets allow this) Opus will be implemented for recording in the reasonably near future, though.
04:01:26[Saint]There's a lot of very bright minds working on it presently.
04:01:33***Saving seen data "./dancer.seen"
04:01:41derf[Saint]: Excellent.
04:01:42[Saint](Not Rockbox Opus encoding, but, Opus in general)
04:02:12derfCELT encoding complexity is roughly double that of decode.
04:02:20the-kyleI think I had mentioned Opus as something that would be awesome for recording, but I also thought that decoding was more important. I still do feel this way at present. It's a great thing, but recording can only work if the recording can be played on the device.
04:02:40derfWhich is substantially better than, e.g., Vorbis or lame, where it's roughly 10x.
04:02:46[Saint]Well...not necessarily.
04:03:19[Saint]It would make it a lot less useful, but it would be possible to record a format the device was incapable of playing.
04:03:38[Saint]very silly...but, possible ;)
04:05:32[Saint]derf: I don't follow the codec side of our tree here that much (I think codec guys are all nuts ;)), but I understand you're one of many major players in the Opus world?
04:05:43[Saint](Or, just a very entusiatic onlooker?)
04:05:54derf[Saint]: I wrote about a third of the codec.
04:06:09derfAnd over half the specification.
04:06:26*[Saint] read "speculation" there for a second...
04:07:33 Quit froggyman (Ping timeout: 246 seconds)
04:08:14 Quit tjb0607 (Ping timeout: 256 seconds)
04:08:51[Saint]I haven't been using Opus on my DAPs, but I transcoded a bunch of flac to opus 128k/s and chucked it on my phone, and all is well there.
04:09:15[Saint]From my understanding, presently, most of my DAPs should be able to do 128k in real-time now, though.
04:11:07[Saint]I need to put some numbers of for MSM7227(T) on the codec comparison chart we have, my numbers blew away the numbers up there at the moment.
04:12:11derfWhich codec comparison chart is this?
04:13:02[Saint]derf: there's some very old numbers here, but some reasonably new numbers also:
04:13:03 Join tjb0607 [0] (~tjb0607@
04:13:36the-kyle[Saint]: What are you using to play Opus on your phone? Is this in rockbox dev build?
04:13:49[Saint]the-kyle: correct.
04:14:43*the-kyle needs to try that again, but couldn't make it talk, and I don't think it's working with hardware keyboards at present.
04:15:26[Saint]derf: those numbers are generated with the set of test files we have (I'm not actually sure if we have opus test files yet? maybe I'll add some later if no one beats me to it...) and the (compile-time optional) test_codec plugin.
04:16:09[Saint]the-kyle: RaaA cannot do speech, a filesystem/permissions thing, iiuc.
04:16:51[Saint]Also, it _should_ work fine with a hardware keyboard.
04:17:19derfWow, is AAC-HE really that much more expensive than AAC-LC?
04:17:41[Saint]We don't use the crazy-ass Rockbox virtual keyboard, but Android's native text entry magic afaik, so, it should "Just Work"
04:17:46[Saint]the-kyle: ^
04:17:53[Saint]derf: yeppers.
04:18:03the-kyleOK. I think it was just the fact that I couldn't get voice/talk clips working. I really wasn't able to test much because Rockbox had Laryngitis.
04:19:06derfI thought Opus was roughly equal in complexity to HE-AAC (but never had an implementation to test with). I guess I know better now.
04:19:08[Saint]derf: the numbers can vary pretty weirdly between various targets as well.
04:19:20derfWell, I looked at more than one.
04:19:31 Quit nateloaf (Read error: Connection reset by peer)
04:19:58[Saint]the-kyle: Oh, ....are you (partially) blind?
04:21:17[Saint]If so, I now understand the hardware keyboard and need for voicing.
04:21:32[Saint](and why RaaAoA woukld be totally useless to you)
04:23:18 Quit speckmade1 (Ping timeout: 246 seconds)
04:23:50 Join speckmade [0] (
04:29:55 Join eckoit [0] (~ryan@
04:32:08 Quit TheSphinX^ (Read error: Operation timed out)
04:33:12 Quit eckoit (Client Quit)
04:38:58 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:38:59 Quit amiconn (Disconnected by services)
04:39:01 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:39:10 Join TheSphinX^ [0] (
04:39:11 Quit pixelma (Disconnected by services)
04:39:11 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:39:13 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:41:10 Quit ps-auxw (Ping timeout: 272 seconds)
04:42:08 Join eckoit [0] (~ryan@
04:50:54 Join saratoga [0] (123e0cfc@gateway/web/freenode/ip.
04:51:15saratogaAAC-HE is insanely slow because of the filterbanks needed to serperate the spectral bands
04:51:53 Join ps-auxw [0] (
04:52:04saratogafrom what i've seen, Opus should be more like AAC-LC in terms of complexity
04:52:37derfI had just assumed that SBR couldn't be much more complex than the folding stuff we do.
04:53:14derfAnd AAC-LC doesn't have anything like the TF-resolution changes we have, AFAIK.
04:53:55saratogafrom the profile results I've seen, it seemed like most of the codec runtime was still the MDCT on ARM
04:54:10saratogawhich is the case for most transform codecs
04:54:10derfYes, that is definitely the most expensive part.
04:54:55saratogaon AAC-HE you have to run a couple filterbanks (i forget the details, but basically a lot of QMF stuff) and that is insanely slow on ARM
04:55:07saratogaas well as the MDCT in the LC part of the codec
04:55:33derfI guess I don't understand why they didn't just split the MDCT coefficients up into bands like we do.
04:55:51saratogathey need to extract different subbands
04:55:58saratogathe core of the codec is still AAC-LC
04:56:17saratogathey take the LC decoded stream and then feed that into kind of a weird subband type codec
04:56:33derfThat sounds awful.
04:56:33saratogai guess whoever came up with the MP3 hybrid filterbank still works for MPEG :)
04:56:39derfHah, yes.
04:57:04saratogabut yes, if you profile LC, you spend pretty much all your time in the analysis and synthesis filterbanks
04:57:21saratogai forget exactly how they fit together, but they each use more then the MDCT in AAC-LC IIRC
04:57:26saratogasorry, profile HE
04:57:44derfThat's incredible.
04:58:03derfI'm glad I didn't waste any time looking at how it was designed.
05:00:12saratoga <- some info
05:01:22the-kyle[Saint]: I'm mostly blind as a matter of fact. Light and very high contrast stuff mostly. Letters need to be light on dark background and insanely large for me to see them.
05:01:38 Part eckoit
05:01:55saratogaactually i don't have a good idea what problem HE is trying to solve
05:02:18derfThe problem it is trying to solve is "we don't have enough bits to code anything in the HF".
05:02:22saratogai guess the need for different time/frequency resolution at different bands combined with trying to squeeze out more compression by not really encoding high frequencies?
05:03:15derfWe do a similar "split things into different pieces" in Hybrid mode, where SILK takes the LF and CELT the HF.
05:03:24derfBut we don't have expensive filters in the decoder.
05:03:39derfYou literally just decode each half and add the two signals together.
05:04:35derfBut back to SBR...
05:04:37saratogai guess they do it that way so that HE can be backward compatible with LC
05:05:33derfOne of the big issues with stuff like MP3 is that coding a _single_ non-zero coefficient in one of the HF bands is very expensive.
05:05:46 Quit speckmade (Ping timeout: 244 seconds)
05:05:53derfBecause you've got, for example, a few hundred of them.
05:06:27derfSo you're talking 7 or 8 bits of entropy just to say which one it is.
05:07:04derfSo what happens is you use a very coarse quantizer, and most of the time all or almost all of the coefficients fall under that threshold.
05:07:22derfHold on, brb.
05:10:23 Quit TheSeven (Disconnected by services)
05:10:32 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:11:26derfOkay, back.
05:11:54derfSo, anyway, what happens is you really have a bunch of low-energy coefficients spread around that band.
05:12:04derfAnd occasionally one of them will get over the quantization threshold.
05:12:25derfAnd so you'll code a non-zero value in that one spot, and then the next frame it will be under the threshold and disappear again.
05:12:45derfOf course, a single non-zero value sounds very tonal, even if the original signal wasn't tonal at all.
05:12:56derfSo you get these tweeting "birdie" artifacts.
05:13:13derfAnd every time you add one, you spend a bunch of bits to do it.
05:13:34derfMake sense so far?
05:13:47 Join froggyman [0] (~froggyman@unaffiliated/froggyman)
05:21:20 Quit ukleinek (Ping timeout: 256 seconds)
05:21:25 Quit Clear_runway (Ping timeout: 245 seconds)
05:34:42 Quit XavierGr (Ping timeout: 244 seconds)
05:37:56 Quit saratoga (Ping timeout: 245 seconds)
05:42:05[Saint]the-kyle: that's...unfortunate. I could probably knock up a trivial blind/high contrast touchscreen theme for RaaA without too much effort, and it's probably something we should have in the source tree...
05:42:44[Saint]Ideally, voicing should work on Android, but even then you'd need a .wps you could use easily.
05:42:45 Join ukleinek [0] (
05:43:00 Join amayer [0] (
05:43:55[Saint]How I'm thinking of doing it, I should be able to make it work for any given resolution as well I think.
05:49:24 Quit ender| (Ping timeout: 246 seconds)
05:51:53the-kyle[Saint]: In order for blind users to use the touch screen functions, some voicing would need to be done to allow exploring the screen with a finger, i.e. dragging a finger around the screen with each icon being spoken as the finger lands on it, similar to the way Android 4.0 with explore by touch works. Another option is something similar to the eyes-free shell, which is arranged similar to a 3 by 3 grid, with the place where the finger l
05:52:37the-kyleThere are also other options, and I would be happy to test different things.
05:53:07the-kyleI think I have all I need to build RaA for Android.
05:54:37the-kyleOf course, being able to use a hardware keyboard if a device has it is also good.
05:54:58 Quit froggyman (Quit: ZNC -
06:01:36***Saving seen data "./dancer.seen"
06:03:05 Join ender| [0] (~ender1@2a01:260:4094:1:42:42:42:42)
06:22:41 Join mystica555 [0] (
06:23:22 Join Braniac [0] (~d4rkc0d3@
06:28:31 Quit Braniac (Quit: Saliendo)
06:39:33 Join froggyman [0] (~froggyman@unaffiliated/froggyman)
06:44:58 Quit crwl (Ping timeout: 244 seconds)
06:46:43 Join crwl [0] (
07:03:07 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
07:05:04 Join Keripo [0] (
07:05:11 Nick PorkerR is now known as DoctorWho (ParkerR@unaffiliated/parkerr)
07:05:18 Nick DoctorWho is now known as ParkerR_ (ParkerR@unaffiliated/parkerr)
07:13:34 Part amayer
07:46:47 Join mortalis [0] (
07:48:16[Saint]JdGordon: ....hey, old buddy, old pal, honey hard would it be to check for the existence of an arbitrary file in the skin code? ;)
07:48:27[Saint]*with a skin tag, rather
07:48:43JdGordonpretty easy
07:48:51JdGordon<10 LOC
07:49:23JdGordonpresumably this is asking if AA exsists for the track?
07:49:31JdGordonor lyrics?
07:50:21[Saint]lyrics was one thing I thought of, but I was thinking it'd be a reasonably easy way for an author to give a user to override custom fonts.
07:50:25 Quit Raptors (Read error: Connection reset by peer)
07:50:45 Join Raptors [0] (~Raptors@
07:51:02JdGordonwell yeah, its pretty simple to do
07:51:27[Saint]check if "custom_font.fnt" exists, and fire a viewport that uses it instead of forcing the font that's hardcoded.
07:51:40[Saint]for example.
07:52:00JdGordonbetter would be conditional font loading so the numbers would be the same
07:52:03JdGordonbut not really possible
07:52:06JdGordongimme 5 min
07:52:16[Saint]You don't have to do it *now* ;)
07:52:29[Saint]...or, at all, for that matter.
07:52:51JdGordontag id?
07:53:36[Saint]it kinda sucks, but, is %fp taken? (file present?)
07:53:57JdGordonthats file path atm
07:54:15JdGordonpath existst?
07:54:22JdGordonno, thats gone too
07:54:28 Quit KiwiCam (Ping timeout: 265 seconds)
07:54:51[Saint]%pp? (path present)
07:55:00[Saint]I have a feelign that's gone as well...
07:55:16JdGordonxx for the time being
07:56:06[Saint]I guess you could read that as "x exists?", good enough :)
07:57:07*[Saint] wonders if skin tags will ever reach a critical point when we start using 3+ digits or full strings.
07:57:12 Join KiwiCam [0] (~quassel@
07:57:23[Saint]the latter would be nice to read.
08:01:39***Saving seen data "./dancer.seen"
08:03:53 Quit ps-auxw (Ping timeout: 260 seconds)
08:06:30JdGordon[Saint]: g#334
08:06:32fs-bluebotGerrit review #334 at : skin_engine: new tag %xx to see if a file exists... %xx(/path/to/file) by Jonathan Gordon (changes/34/334/1)
08:06:37JdGordoncompltly untested or built :)
08:06:40JdGordonand not finished
08:06:44JdGordonbut enough to get you going
08:06:52[Saint]awesome, thanks.
08:07:20JdGordonit only works with full paths now, but it would need to work with the current files path
08:07:25JdGordonand possibly next files path too
08:07:42[Saint]ah, yes.
08:15:08 Join ps-auxw [0] (
08:15:09 Quit KiwiCam (Ping timeout: 245 seconds)
08:17:54 Join KiwiCam [0] (~quassel@
08:18:49 Join speckmade [0] (
08:25:02 Quit KiwiCam (Read error: Operation timed out)
08:28:05 Join KiwiCam [0] (~quassel@
08:35:28 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:37:46 Quit speckmade (Remote host closed the connection)
08:40:19 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:50:59 Join ender` [0] (
08:58:17 Quit Galois (Ping timeout: 272 seconds)
09:06:25 Quit Keripo (Read error: Connection reset by peer)
09:09:47 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:14:09 Join LinusN [0] (
09:14:31pamaurywodz (log): do you know if RAThomas comes regularly on the chat ? I know he has a rknano device and I want to ask him something but he not come recently I think
09:27:15 Quit bertrik (Ping timeout: 256 seconds)
09:36:36 Join wodz [0] (
09:37:05wodzpamaury: don't know
09:37:10 Quit kevku (Ping timeout: 260 seconds)
09:37:55 Join kevku [0] (
09:56:06 Quit wodz (Quit: Leaving)
09:57:16 Join fml [0] (
09:57:32fmlbluebrother: to review: g#330
09:57:33fs-bluebotGerrit review #330 at : Add missing commas in German translation of rbutil by Alexander Levin (changes/30/330/1)
09:57:37 Quit fml (Client Quit)
10:01:41***Saving seen data "./dancer.seen"
10:05:06 Join GodEater_ [0] (93722cc9@rockbox/staff/GodEater)
10:40:52 Join wodz [0] (
10:51:43 Join Keripo [0] (
10:58:41 Join Galois [0] (
11:06:06 Quit GodEater_ (Quit: Page closed)
11:13:23 Quit pamaury (Remote host closed the connection)
11:14:58 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:16:53 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
11:16:58 Quit Keripo (Quit: Leaving.)
11:19:19 Quit pamaury (Remote host closed the connection)
11:20:00 Join pamaury [0] (
11:20:01 Quit pamaury (Changing host)
11:20:01 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:20:43 Quit pamaury (Remote host closed the connection)
11:26:04 Quit dys (Read error: Connection reset by peer)
11:26:30KiwiCamSo, how could I make use of this new tag xx%? What would you use it for?
11:26:58wodzkugel, others: <- What do you think about such buffers management? This will almost automatically add overlay capability to all plugins. The plugin side needs to be check but this should be easy.
11:27:06wodzAm I missing something?
11:27:46 Join dys [0] (~user@2a01:1e8:e100:8296:21a:4dff:fe4e:273a)
11:28:02wodzbtw. It this is ok we could merge plugin_get_buffer() and plugin_get_audio_buffer() into single one with additional parameter which buffer you would like to get
11:33:05 Join |akaWolf| [0] (~akaWolf@
11:33:46 Quit akaWolf (Ping timeout: 240 seconds)
11:35:20kugelwodz: what does it do exactly?
11:36:41 Join pamaury [0] (
11:36:41 Quit pamaury (Changing host)
11:36:41 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:37:39wodz1) if elf_open() fails due to lack of mem in pluginbuf it grabs audiobuf and tries again in there. 2) it changes plugin_get_audio_bufer() to be like plugin_get_buffer() e.g it checks if plugin is loaded into audiobuf and if so it accounts for that in returned pointer and size
11:38:16 Join speckmade [0] (
11:50:07 Join burncode [0] (
11:50:45burncodeHello everybody!
11:52:52burncode"android-ndk-r8b/platforms/android-5/arch-arm/usr/lib/crtend_so.o: Unknown EABI object attribute 44" I compiled Android version of the time
11:53:32burncodeKnow what causes it
11:53:34burncodeKnow what causes it
11:54:09 Quit bluebrother (Disconnected by services)
11:54:14 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
11:54:26burncodeIs there anyone here
11:54:41 Join lebellium [0] (
11:55:03burncodeIs there anyone here
11:56:43[Saint]please don't spam our channel...
11:56:54[Saint]it really doesn't make anyone want to help.
11:57:15 Quit fs-bluebot (Ping timeout: 246 seconds)
11:58:31 Join fs-bluebot [0] (
12:00:06 Quit burncode (Quit: CGI:IRC (Ping timeout))
12:01:36[Saint]what is up with that string anyway...?
12:01:42***Saving seen data "./dancer.seen"
12:02:19[Saint]oh, nevermind.
12:06:28wodzkugel: any comments/thoughts/remarks?
12:32:06 Quit speckmade (Ping timeout: 252 seconds)
12:38:48wodzgevaerts: which combination of binutils/gcc is used to build rockbox for MIPS/android?
12:41:59 Join petur [0] (~petur@rockbox/developer/petur)
12:57:38[Saint]wodz: as far as I understand it, it is handled simply by the SDK automagically by passing an additional flag.
12:59:16wodz[Saint]: ? Doesn't android sdk use particular combo?
13:02:38[Saint]I'm unsure if it's entirely different, or if you just need to use the MIPS versions of S/NDK
13:04:19 Join minouch [0] (
13:04:24[Saint]Oh, there's no such thing anymore..apparently. It seems it's all been combined.
13:06:45kugelwodz: sounds good
13:07:10kugelwodz: android uses the ndk toolchains
13:08:47wodz... and ndk gcc/binutils combo is?
13:10:25[Saint]gcc == arm-linux-androideabi-gcc 4.4.3 (404)
13:11:22gevaertswodz: seems to be gcc 4.4.3 and binutils 2.21
13:11:38wodzok, thanks
13:11:41[Saint]bunutils == 2.21
13:11:46[Saint]ah, too late ;)
13:15:58[Saint]KiwiCam: as mentioned earlier, it could be used for: 1 - a "magic file" based configuration system for theme elements, 2 - checking to see if a given file is present in order to use it to override a hardcoded default, 2 - checking to see if a bookmark/lyrics/album art exists...etc.
13:16:21[Saint]the case to override a hardcoded default was what I was interested in particularly.
13:17:04[Saint]the theme could check for the presence of "custom_font1.fnt" and fire a viewport that loads and uses it, instead of the fall-through hardcoded case, for example.
13:17:37[Saint](not that file names used are for example, and any path/extension could be checked for existence.
13:17:46wodzbtw. I discovered a few bugs in jz4740 lowlevel code when I was playing with qemu-jz. Is it ok to just commit fixes or should I try to prompt Maurus?
13:17:47[Saint]*note that
13:18:25wodzI have no way to test other then matching qemu-jz behavior and docs.
13:18:57gevaertsI think we need to either find someone who can test on mips or drop those players
13:19:04[Saint]KiwiCam: for the "magic file" based configuration case, things could get really interesting. You could make conditional cases that entirely changed the layout or elements presented in the theme by checking for the presence of one or more "magic files"
13:19:19wodzgevaerts: or by one to the pool
13:19:53wodzalthough I recall my mail about this left unanswered
13:20:00[Saint]KiwiCam: you could, in theory, allow a user to turn on or off any theme element used in the theme by adding or removing a magic file.
13:20:06gevaertsYes, that's definitely an option too
13:20:28[Saint]this would make the conditional checks in the theme reasonable complicated, but, it would be *very* awesome (IMO)
13:21:30*[Saint] is getting a bunch of:
13:21:35[Saint]"LD test_mem.rock
13:21:36[Saint]/home/saint-ubuntu/Development/Android/android-ndk-r8b/toolchains/arm-linux-androideabi-4.4.3/prebuilt/linux-x86/bin/../lib/gcc/arm-linux-androideabi/4.4.3/../../../../arm-linux-androideabi/bin/ld: Warning: /home/saint-ubuntu/Development/Android/android-ndk-r8b/platforms/android-5/arch-arm/usr/lib/crtend_so.o: Unknown EABI object attribute 44" in git HEAD
13:22:25wodzit seems like still have some vx747 on stock
13:22:33[Saint]it seems to be all the codec and rock files...perhaps wodz angered some warning beast? ;)
13:23:21[Saint]Hmmmm, yes. It is, it's all codecs and rocks spitting that warning.
13:23:49[Saint]it appears to still compile fine, so possibly just needs some magic to silence it.
13:32:00 Join XavierGr [0] (XavierGr@rockbox/staff/XavierGr)
13:34:04 Join TorneWuff [0] (
13:38:28 Quit jm (*.net *.split)
13:38:28 Quit Torne (*.net *.split)
13:38:29 Nick TorneWuff is now known as Torne (
13:38:29 Quit Torne (Changing host)
13:38:29 Join Torne [0] (~torne@rockbox/developer/Torne)
13:41:45wodzgcc-4.6.3 found error in libgme cpp conditions :-)
13:59:53 Quit [Saint] (Ping timeout: 246 seconds)
14:00:02 Join [Saint] [0] (~quassel@rockbox/user/saint)
14:01:45***Saving seen data "./dancer.seen"
14:07:39 Join dfkt [0] (dfkt@unaffiliated/dfkt)
14:13:42 Quit wodz (Quit: Leaving)
14:17:31 Join amayer_ [0] (
14:49:27 Quit minouch (Quit: CGI:IRC)
14:59:31 Join maruk [0] (
14:59:44 Part maruk
15:17:23 Join WalkGood [0] (~4@unaffiliated/walkgood)
15:24:05 Join GeekShadow [0] (
15:24:05 Quit GeekShadow (Changing host)
15:24:05 Join GeekShadow [0] (~antoine@reactos/tester/GeekShadow)
15:37:57 Join jm [0] (
15:41:29 Part LinusN
15:45:20 Quit mortalis (Quit: Leaving)
16:01:49***Saving seen data "./dancer.seen"
16:06:06 Quit tjb0607 (Ping timeout: 268 seconds)
16:07:03 Join tjb0607 [0] (~tjb0607@
16:19:28 Quit pamaury (Read error: Operation timed out)
16:30:27 Join Strife89 [0] (~Strife89@
16:42:25 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
16:43:07 Quit petur (Quit: *plop*)
16:43:09 Join Rower85 [0] (
16:59:16funman : lcd_flip on clip zip
16:59:23funmanthere are some glitches when flipped though
17:08:58 Join pretty_function [0] (~sigBART@
17:11:26 Join Poodlemastah [0] (
17:14:30 Quit mgottschlag (Ping timeout: 260 seconds)
17:24:55 Quit Zagor (Quit: Clint excited)
17:26:11 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
17:34:54 Quit WalkGood (Ping timeout: 268 seconds)
17:37:29funmanmaybe bertrik will know what's going on
17:37:53 Join nateloaf [0] (
17:38:23 Join WalkGood [0] (
17:42:04 Quit pretty_function (Remote host closed the connection)
18:01:50***Saving seen data "./dancer.seen"
18:14:17 Quit FOAD (Quit: I'll be back)
18:14:36 Join FOAD [0] (~foad@
18:16:09 Quit WalkGood (Read error: Connection reset by peer)
18:16:43 Join WalkGood [0] (
18:16:44 Quit WalkGood (Changing host)
18:16:44 Join WalkGood [0] (~4@unaffiliated/walkgood)
18:35:51 Quit WalkGood ()
18:43:28 Quit amayer_ (Quit: going ~/)
18:45:30 Quit tjb0607 (Ping timeout: 260 seconds)
18:47:33 Join tjb0607 [0] (~tjb0607@
18:55:29 Join amayer_ [0] (
19:03:53 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
19:12:36 Quit efyx_ (Ping timeout: 245 seconds)
19:16:08 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:20:08nrghats off to you guys, just used the rockbox utility to install the latest dev firmware to my new clip zip
19:20:13nrgworked flawlessly
19:25:02 Quit hype (Quit: ["Textual IRC Client:"])
19:26:02 Join efyx_ [0] (~efyx@
19:26:35 Quit tjb0607 (Ping timeout: 265 seconds)
19:28:28 Join tjb0607 [0] (~tjb0607@
19:42:21 Quit thegeek (Read error: Connection reset by peer)
19:44:06 Join thegeek [0] (
19:45:56 Join Horscht [0] (
19:45:56 Quit Horscht (Changing host)
19:45:56 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:46:31 Quit tjb0607 (Ping timeout: 256 seconds)
19:46:44amayer_would anyone be against me adding something to the theme engine to make scroll bars themeable?
20:01:53***Saving seen data "./dancer.seen"
20:05:07 Quit y4n (Disconnected by services)
20:05:14 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
20:05:57 Quit y4n (Read error: Connection reset by peer)
20:06:06 Quit scorche (Disconnected by services)
20:06:10 Join scorche` [0] (~scorche@rockbox/administrator/scorche)
20:07:46funmanbertrik: did you see my message about lcd_flip ?
20:08:48bertrikfunman: no
20:09:03 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
20:09:57bertrikit's fine with me if you implement it for the display of the clip zip you have
20:10:22bertrikyou and I have different kinds of clip zip displays IIRC
20:11:38funmanah you have the type 1 ?
20:14:15 Quit Horscht (Ping timeout: 255 seconds)
20:16:37 Join Horscht [0] (
20:16:38 Quit Horscht (Changing host)
20:16:38 Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:18:13bertrikI don't remember to be honest ...
20:20:23lebelliumyou have the Visionox display bertrik
20:20:50 Join hype [0] (~hype@
20:21:33lebelliumFunman and I have the WiseChip display
20:22:21 Join LinusN [0] (
20:23:34funmanlebellium: can you try to flip LCD with ?
20:25:03 Quit |akaWolf| (Ping timeout: 255 seconds)
20:25:10lebelliumnope sorry, I need a compiled build for that, I'm not a dev and don't compile myself ;)
20:32:45funmanlang files might be a bit off since yours won't have lcd_flip ...
20:35:17funmanlebellium: funman/">
20:35:35lebelliumfunman: just replacing the rockbox.sansa file
20:35:40lebelliumit doesn't work properly
20:36:05lebelliumat least with my theme
20:37:20funmanis there some jumping lines ?
20:37:34lebelliumit just behaves crazy
20:37:37funmani tried demo and they work perfectly as long as i dont browse with the keys
20:39:12lebelliumyou I press "up" or "down" the screen is shaking
20:39:18lebelliumwhen I press*
20:39:44lebelliumshould I install the whole build instead of just rockbox.sansa ?
20:42:26 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
20:49:26 Quit y4n (Read error: Connection reset by peer)
20:50:12lebelliumfunman: the same with the whole build you provide. the LCD is flipped but it's shaking when you press a button. With cabbiev2 or my theme, whatever
20:58:22 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
21:01:45 Quit Horscht (Quit: Verlassend)
21:13:31 Quit y4n (Disconnected by services)
21:13:37 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
21:17:35 Quit pamaury (Quit: No Ping reply in 180 seconds.)
21:18:46 Join pamaury [0] (
21:18:46 Quit pamaury (Changing host)
21:18:46 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
21:19:28 Quit Rower85 (Quit: Hmmm...)
21:19:38 Join Horscht [0] (
21:19:38 Quit Horscht (Changing host)
21:19:38 Join Horscht [0] (~Horscht@xbmc/user/horscht)
21:20:46 Quit y4n (Disconnected by services)
21:20:52 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
21:40:46 Quit Gallomimia (Quit: I am likely going to change locations)
21:44:00 Quit mgottschlag (Read error: Connection reset by peer)
21:55:26 Join Bagder_ [0] (~daniel@
21:56:20 Join petur [0] (~petur@rockbox/developer/petur)
22:01:54***Saving seen data "./dancer.seen"
22:02:39 Join MasT3R [0] (
22:03:57 Quit y4n (Disconnected by services)
22:04:04 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
22:07:45 Quit y4n (Disconnected by services)
22:08:02 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
22:08:06 Quit y4n (Client Quit)
22:16:14 Quit Provel (Read error: Connection reset by peer)
22:16:39 Join Provel [0] (
22:26:49 Quit alexbobp (Ping timeout: 252 seconds)
22:42:05 Quit shamus (Read error: Connection reset by peer)
22:42:13 Join shamus [0] (
22:45:38 Join alexbobp [0] (
22:47:33 Quit Bagder_ (Remote host closed the connection)
22:59:29 Part LinusN
23:01:17 Join Clear_runway [0] (~Clear_run@
23:03:37 Quit MasT3R (Quit: CGI:IRC (EOF))
23:05:31 Quit amayer_ (Ping timeout: 256 seconds)
23:06:23 Quit Strife89 (Quit: Heading out)
23:14:26 Join mgottschlag [0] (~quassel@reactos/tester/phoenix64)
23:29:53 Quit mgottschlag (Ping timeout: 246 seconds)
23:32:17 Quit kevku (Ping timeout: 272 seconds)
23:42:01 Join MasT3R [0] (
23:49:47 Quit petur (Remote host closed the connection)
23:53:17 Quit MasT3R (Quit: CGI:IRC (EOF))
23:54:48 Quit alexbobp (Quit: migrating)

Previous day | Next day