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

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

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

#rockbox log for 2013-07-06

00:08:17 Quit pamaury (Ping timeout: 264 seconds)
00:23:08 Join Raptors_ [0] (~whoneedsa@216-58-33-203.cpe.distributel.net)
00:23:08 Quit Raptors (Disconnected by services)
00:25:25 Quit petur (Quit: Leaving)
00:39:35 Quit Zarggg (Ping timeout: 276 seconds)
00:39:39 Quit y4n (Quit: AMIGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHAHAHAAAAAAAAAAAAHAHAAA)
00:43:03 Quit Guest79504 (Read error: Connection reset by peer)
00:47:19 Join nosa-j [0] (~m00k@184.76.254.130)
01:00
01:03:38 Quit ender1 (Quit: Religion is an insult to human dignity. With or without it you would have good people doing good things and evil people doing evil things. But for good people to do evil things, that takes religion. -- Steven Weinberg)
01:12:31 Join mrtux [0] (~mrtux@unaffiliated/mrtux)
01:14:59 Join Zarggg [0] (~zarggg@24.229.140.62.res-cmts.sm.ptd.net)
01:32:05 Quit belak (Ping timeout: 268 seconds)
01:50:36***Saving seen data "./dancer.seen"
01:53:51 Quit Rower (Quit: asdöl)
01:57:40 Quit lebellium (Quit: ChatZilla 0.9.90 [Firefox 23.0/20130703181823])
02:00
02:07:20 Quit bertrik (Ping timeout: 276 seconds)
02:40:11 Quit prof_wolfff (Ping timeout: 256 seconds)
02:42:25jhMikeSsaratoga: I'd consider it essentially ready to push
02:46:36saratogacool, then go for it :)
03:00
03:02:33jhMikeSI suppose I will quite shortly then :)
03:50:40***Saving seen data "./dancer.seen"
03:58:30 Quit Guest46062 (Read error: Connection reset by peer)
03:59:14 Join Guest46062 [0] (~liar@clnet-p09-185.ikbnet.co.at)
04:00
04:55:56 Quit pixelma (Disconnected by services)
04:55:56 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:55:59 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:56:12 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:56:12 Quit amiconn (Disconnected by services)
04:56:14 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
05:00
05:14:58 Quit TheSeven (Disconnected by services)
05:15:08 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:34:53 Quit gelraen (Ping timeout: 264 seconds)
05:40:38 Join gelraen [0] (~imax@mail.univua.net)
05:45:45 Quit Epicanis (Quit: byebye time)
05:50:44***Saving seen data "./dancer.seen"
05:58:46 Join belak [0] (~belak@facebook/engineering/belak)
06:00
06:28:15 Join Guest79504 [0] (uid4289@gateway/web/irccloud.com/x-yyxrzdenxfgqhupt)
06:40:27 Join TeruFSX [0] (~TeruFSX@65-128-175-59.mpls.qwest.net)
06:41:13 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
07:00
07:24:53 Quit Marex_ (Ping timeout: 248 seconds)
07:29:59 Quit pamaury (Ping timeout: 245 seconds)
07:30:44 Join Marex [0] (~Marex@195.140.253.167)
07:41:29 Quit Marex (Ping timeout: 264 seconds)
07:47:20 Join Marex [0] (~Marex@195.140.253.167)
07:50:47***Saving seen data "./dancer.seen"
08:00
08:06:06 Nick funman is now known as craco (~fun@rockbox/developer/funman)
08:06:13 Nick craco is now known as crado (~fun@rockbox/developer/funman)
08:24:18 Join n1s [0] (~n1s@nl118-168-30.student.uu.se)
08:24:18 Quit n1s (Changing host)
08:24:18 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:29:37 Quit bluebrother^ (Ping timeout: 240 seconds)
08:31:54 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
08:34:30 Join kevku [0] (~kevku@2001:0:c38c:c38c:1837:ca71:3d69:befe)
08:56:31 Join EstebanMonge [0] (~emonge@ip56-89-15-186.ct.co.cr)
08:57:05EstebanMongeHello I made a fmr preset file for Costa Rica
08:57:09EstebanMongeI want upload it
08:57:54EstebanMongebut need permissions for Action Change in the wiki
09:00
09:00:32 Join kaputnik_ [0] (~kaputnik@port-92-206-105-85.dynamic.qsc.de)
09:15:54 Quit nosa-j (Excess Flood)
09:16:02 Join nosa-j [0] (~m00k@184.76.254.130)
09:19:32 Quit Guest46062 (Ping timeout: 268 seconds)
09:21:22 Join Guest46062 [0] (~liar@clnet-p09-185.ikbnet.co.at)
09:30:31 Quit Guest46062 (Read error: Connection reset by peer)
09:36:31 Join olspookishmagus [0] (~pookie@91.132.63.143)
09:44:26 Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
09:50:51***Saving seen data "./dancer.seen"
09:53:18 Quit EstebanMonge (Read error: Connection reset by peer)
10:00
10:18:48 Join ender` [0] (krneki@foo.eternallybored.org)
10:21:43 Nick crado is now known as funman (~fun@rockbox/developer/funman)
10:37:27 Quit thegeek (Read error: Connection reset by peer)
10:39:24 Join thegeek [0] (~thegeek@40.200.16.62.customer.cdi.no)
10:44:57 Join einhirn [0] (~Miranda@p4FF8937A.dip0.t-ipconnect.de)
10:45:14 Join lorenzo92 [0] (~chatzilla@host107-109-dynamic.0-87-r.retail.telecomitalia.it)
10:49:09 Quit einhirn (Ping timeout: 248 seconds)
10:58:25 Nick SuperBrainAK is now known as DormantBrain (~andy@shared02.balt01.cd.2g2u.net)
11:00
11:00:09 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:05:26 Join einhirn [0] (Miranda@bsod.vpn.tu-clausthal.de)
11:14:45 Quit einhirn (Ping timeout: 256 seconds)
11:15:01 Quit copper (Read error: Connection reset by peer)
11:29:46 Join prof_wolfff [0] (~prof_wolf@62.83.50.196.dyn.user.ono.com)
11:32:17 Join copper [0] (~copper@unaffiliated/copper)
11:40:09 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
11:45:53 Quit flugfafe1 (Quit: leaving)
11:48:23 Join Rower [0] (husvagn@v-413-alfarv-177.bitnet.nu)
11:50:54***Saving seen data "./dancer.seen"
11:53:29 Quit prof_wolfff (Ping timeout: 264 seconds)
11:55:31 Join prof_wolfff [0] (~prof_wolf@62.83.50.196)
12:00
12:09:43lebelliumI don't quite understand the new 48000Hz patch. Why would someone set it to 48k manually except if he only has 48k files on his player? And if the player hw supports 48k, wouldn't it better to automatically set the frequency depending on the file frequency? And does that mean 24/192 could also be natively supported on "audiophile" targets?
12:14:17[Saint]I would say the commit message answers all of those questions.
12:14:46[Saint](except for the "wouldn't it be better to switch dynamically?" one)
12:15:22lebelliumI read the commit message
12:15:32lebelliumI read it again, maybe I missed something
12:16:43lebelliumso the answer to the last question is "On particular hardware where very high rates are practical and desireable, the upper bound can be extended by patching. ", right?
12:16:58[Saint]yes.
12:17:39lebelliumBut I still find the commit message unclear about why would people need this setting, especially if it has to be selected manually
12:17:42 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
12:17:55[Saint]recording.
12:19:13[Saint]and, to set the minimum rate at which it will be downsampled to on playback (if higher).
12:20:33 Quit JdGordon (Ping timeout: 264 seconds)
12:20:47lebelliumhum okay
12:25:46 Quit JdGordon_ (Ping timeout: 268 seconds)
12:25:55*bertrik looks for wodz
12:26:08 Join JdGordon [0] (~jonno@CPE-58-166-106-40.lnse5.cht.bigpond.net.au)
12:26:09 Quit JdGordon (Changing host)
12:26:09 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:26:23*[Saint] suggests a collar/leash
12:27:17[Saint]"Do *you* know where your developer is? Always respect leash laws."
12:29:05lorenzo92okay the lock-in-FM possibility is buggy: why the hell can someone turn off a player while being in a locked state?!
12:30:49kugel[Saint]: recording?
12:30:53lorenzo92ah okay lebellium reports this fault also being in WPS and locked ... hum ...
12:30:58kugelthe commit is about playback frequency
12:30:59copperlebellium: for stuff like Opus, which is always at 48kHz
12:31:10copperwhich is what motivated saratoga for the path
12:31:15copperpatch*
12:31:42copper(IIRC)
12:32:52lebelliumcopper: but you would not switch the setting everytime you skip from a 44k to a 48k file and the reverse. Unless you only have 48k opus files on your player. That's why I don't really see the point unless it doesn't automatically set the right frequency
12:33:14copperyes, users might only have opus files, to save space
12:33:18[Saint]isn't NATIVE_FREQUENCY used to define the max for recording as well?
12:33:51kugeli dont know
12:34:04kugellebellium: I too wonder why it can't be switched automatically
12:34:33kugeljhMikeS: you're lazy :) the langtool can update all .lang files in a batch
12:34:44copperbecause of the fade-in/out effect?
12:35:14[Saint]but you'd know it was coming.
12:35:58[Saint]unless the buffer only held one track, or part thereof, in which case it wouldn't matter.
12:36:15 Quit JdGordon (Ping timeout: 268 seconds)
12:36:22lorenzo92can someone commit this: http://gerrit.rockbox.org/r/#/c/504/ ... it should be pretty trivial :)
12:38:35[Saint]errrr...hum. no. that doesn't work. no idea what I was thinking/trying to say there.
12:39:24lorenzo92kugel: thanks :)
12:39:37 Join JdGordon [0] (~jonno@101.174.63.168)
12:39:37 Quit JdGordon (Changing host)
12:39:37 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:39:39[Saint]fade in/out _shouldn't_ be affected, if the two are at different sample rates...I don't think.
12:39:43kugelyou're welcome :)
12:42:29lebelliumif there are no drawbacks such as fade in/out effects, I vote for automatic switch. Yes I know, easier to say than to do :D
12:44:17 Quit JdGordon (Ping timeout: 256 seconds)
12:46:28 Join fyre^OS [0] (~nnscript@cpe-24-90-158-3.nyc.res.rr.com)
12:49:53 Quit fyrestorm (Ping timeout: 264 seconds)
12:50:07 Join JdGordon [0] (~jonno@CPE-58-164-36-57.lnse5.ken.bigpond.net.au)
12:50:07 Quit JdGordon (Changing host)
12:50:07 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
12:57:42kugellorenzo92: I havent checked kubuntu 13.04, but *buntu is known to break toolchain builds every now and then
13:00
13:07:06 Quit jhMikeS (Ping timeout: 276 seconds)
13:09:51 Join einhirn [0] (~Miranda@p4FF8937A.dip0.t-ipconnect.de)
13:09:58 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
13:12:21lorenzo92kugel: ok good to know ;)
13:36:21 Join webguest033 [0] (~d9ee9ab3@www.haxx.se)
13:36:45 Quit webguest033 (Client Quit)
13:36:46 Join ZincAlloy [0] (~d9ee9ab3@www.haxx.se)
13:46:50lorenzo92http://gerrit.rockbox.org/r/#/c/505/ : bugfix solved on R0, now lebellium tries on other devices
13:50:55***Saving seen data "./dancer.seen"
13:56:53 Quit Galois (Ping timeout: 240 seconds)
13:57:01pamaurylebellium: kugel: I think the reason why we don't automatically switch is because the effect on battery life is unclear. Manually setting the frequency is not the best solution imo but automatic switch may not be either
13:58:03lebelliumlorenzo92: bug also fixed on Clip Zip. Great patch :)
13:58:37lorenzo92lebellium: thanks ;)
13:59:10lebelliumpamaury: ok I did not think about battery life indeed
13:59:19lorenzo92quite simple, but solves a nasty bug ;)
14:00
14:01:46pamaurybut clearly this is a good step in the good direction: now we can change the dsp frequency !
14:11:16lorenzo92I'm wondering how to mess with a cramfs image in a windows system. I would only need to insert some files in it, because I'm starting thinking about YP-R0 and rbutil again
14:21:02 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
14:21:17[Saint]The issue isn't how to do it, the issue is how to do it with tools with permissive licenses.
14:22:51kugelpamaury: only jhMikeS knows the reason i guess
14:23:32lorenzo92"tools with permissive licenses" which ones?
14:23:46lorenzo92we have our unpacker (we can get rid of the muonencrypt)
14:24:59[Saint]My mistake, I was unaware there was a cramfs packer in the source.
14:26:11[Saint]I thought you would have to write one (non-trivial, but highly possible), or find one with a permissive license (quite non-trivial, based on my experience from some years prior).
14:26:37lorenzo92we use the cramfsprogs one i guess
14:26:57lorenzo92and once we get rid of the residual stuff from samsung there should be no further problems
14:27:28lorenzo92i'm now searching a way of putting some files into an image *without* extraction
14:28:24lorenzo92because there is the option "Insert a file image into the filesystem" but it's very unclear ^^
15:00
15:11:19 Quit kevku (Ping timeout: 260 seconds)
15:12:08 Join dfkt [0] (dfkt@unaffiliated/dfkt)
15:23:51 Join kevku [0] (~kevku@2001:0:c38c:c38c:285f:3fc7:3d69:be77)
15:41:09 Quit lorenzo92 (Ping timeout: 276 seconds)
15:43:21 Quit einhirn (Ping timeout: 256 seconds)
15:46:33 Quit ZincAlloy (Quit: CGI:IRC)
15:50:58***Saving seen data "./dancer.seen"
15:56:52 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
15:57:04 Quit pamaury (Ping timeout: 245 seconds)
16:00
16:21:40 Join mrtux_ [0] (~mrtux@unaffiliated/mrtux)
16:23:32 Quit mrtux (Ping timeout: 264 seconds)
16:43:16 Quit dfkt (Remote host closed the connection)
16:54:01 Nick mrtux_ is now known as mrtux (~mrtux@unaffiliated/mrtux)
17:00
17:01:33 Join kaputnik__ [0] (~kaputnik@port-92-206-78-150.dynamic.qsc.de)
17:05:08 Quit kaputnik_ (Ping timeout: 248 seconds)
17:25:24 Quit nosa-j (Ping timeout: 248 seconds)
17:26:49 Join nosa-j [0] (~m00k@184.76.254.130)
17:37:56 Join mirak [0] (~mirak@lns-5-49-104-130.dsl.dyn.abo.bbox.fr)
17:51:00***Saving seen data "./dancer.seen"
17:51:58 Quit olspookishmagus (Ping timeout: 256 seconds)
17:53:01 Join olspookishmagus [0] (~pookie@91.132.63.143)
17:53:25 Nick olspookishmagus is now known as Guest668 (~pookie@91.132.63.143)
18:00
18:08:24 Quit bertrik (Ping timeout: 256 seconds)
18:35:44 Join Strife89 [0] (~Strife89@207-144-5-207.cstel.net)
18:37:51 Quit Guest668 (Quit: All for nothing)
18:52:20 Quit Strife89 (Ping timeout: 248 seconds)
18:54:26 Join Strife89 [0] (~Strife89@2602:306:250e:f6e9:6cc3:cf5e:8daf:6b91)
19:00
19:22:27 Quit alexbobp (Ping timeout: 246 seconds)
19:31:18 Join belak51 [0] (~belak@facebook/engineering/belak)
19:33:12 Quit belak (Ping timeout: 276 seconds)
19:36:12 Join alexbobp [0] (~alex@capitalthree.pwnz.org)
19:39:08 Quit Zarggg (Ping timeout: 264 seconds)
19:40:11 Join dewlap [0] (~dewlap@2001:5c0:1400:a::127)
19:44:56 Nick belak51 is now known as belak (~belak@facebook/engineering/belak)
19:51:03***Saving seen data "./dancer.seen"
19:52:34 Join Zarggg [0] (~zarggg@24.229.140.62.res-cmts.sm.ptd.net)
19:55:44 Join Epicanis [0] (~Epicanis@sdsl-68-238-63-20.static.ngn.east.myfairpoint.net)
20:00
20:02:01copper2012/05/04 21:43:27 UTC <copper> AlexP: it would be nice to at least feed the DAC what it supports, i.e. 48 kHz
20:02:08copper2012/05/04 21:44:13 UTC <AlexP> I don't know the details, but IIUC that would require rewriting large parts that assume 44.1
20:02:12copper2012/05/04 21:44:23 UTC <gevaerts> copper: now implement crossfade on that system :)
20:02:27copper2012/05/04 21:49:05 UTC <saratoga> the conclusion was that it wasn't worthwhile to allow multiple sampling rates
20:02:49copperoh wow, that was a year ago already?
20:04:27gevaertscopper: the thing mentioned earlier was actually fade in/fade out, not crossfade :)
20:05:07*gevaerts will leave these things to those who know the audio bits in rockbox though
20:10:26 Quit shamus (Read error: Connection reset by peer)
20:11:17 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
20:31:31copperisn't it the same code? (I have no clue)
20:33:37 Join petur [0] (~petur@rockbox/developer/petur)
20:50:30 Quit Scall- (Ping timeout: 264 seconds)
20:53:10 Join Scall [0] (~chat@unaffiliated/scall)
20:56:31n1sIIRC the problems with switching automatically were *) audible glitches when changing the rate on some hw *) dsp and decoding changes "tracks" at different times which gets a bit more complicated *) crossfade
20:57:48n1sand i think we already support recording at different samplerates
20:59:46preglowcrossfading basically requires you to lock the sample rate
21:00
21:01:01preglowaudible glitches during silence?
21:01:05preglowthat's pro
21:02:41n1snot sure if they're audible with silence but if you switch at a track change i think jhMikeS said he got glitches on some hw
21:02:54 Quit saratoga (Ping timeout: 250 seconds)
21:03:37preglowwell, tracks with different sample rates should at least not be gapless unless you're doing something you shouldn't
21:03:47preglowso introducing a small silence when changing sample rates would be reasonable
21:04:35n1sand then you get bug reports about gapless being broken etc :)
21:04:54preglowwell, like i said
21:05:03preglowin a gapless album all tracks will be the same sample rate
21:05:27preglowany transitions you notice are broken will be random anyway
21:06:52n1ssure but still, if you listen to a random playlist and if a transition is gapless will depend on the samplerates of the tracks, it will be kind of weird
21:08:32preglowi don't really get it, how can you call a transition between two unrelated tracks "gapless"? they're not supposed to be put together, you'll get some kind of a gap anyway
21:08:39 Quit Strife89 (Quit: Vamoose!)
21:09:10preglowthere's no similarity at the track boundaries that would sound better with or without a gap
21:09:40preglowand in any case, with a bit of luck, the sample rate change discontinuity should always be quite small, 0.1 sec-ish
21:09:59n1suh, i was thinking with crossfade, maybe should have said that ...
21:10:16preglowhahah
21:10:22preglowcrossfade can't use this
21:10:29preglowcrossfade will need to resample
21:10:57 Join saratoga [0] (d8507479@gateway/web/freenode/ip.216.80.116.121)
21:11:34saratogawhen we discussed this a while back jhmikes seemed to think it wasn't worth the complexity in the DSP code, and since very few if any other software platforms can do this, I tended to agree with him that it was impractical
21:11:47saratogahowever now that switching is implemented perhaps he feels differently
21:12:33n1sperhaps we could have a setting for auto switching so users could keep the current behaviour if they wanted
21:14:00saratogawe don't have auto switching at all right now
21:14:12saratogajust a setting for the sampling rate the DSP engine runs at
21:14:33n1syes, but if we add it i think it should be a setting
21:14:57saratogai think the concern was the additional complexity that is involved in making each component of the dsp engine aware of clock change events
21:14:59preglowapparently i need to update my rockbox
21:15:21preglowshouldn't be too much complexity
21:15:40preglowand not having that capability is kind of lame
21:15:56saratogaFWIW, i got interested in better resamplers because i thought it probably wasn't worth trying to change the clock on the fly
21:16:07preglowit's not that big of a deal
21:16:12preglowproblem is just getting a nice enough resampler
21:16:21saratogaif theres still quality problems with our resampler, it should probably be improved IMO
21:17:15saratogai haven't done listening tests, but distortion and aliasing are now massively lower than the old resampler when I tested in matlab
21:17:53gevaertsWhat about voice (and possible other future mixed-in sounds)?
21:21:09saratogai think voice gets resampled if needed
21:21:17saratogacurrently
21:21:38 Quit kevku (Ping timeout: 245 seconds)
21:22:10gevaertsRight, but changing that dynamically could get hairy I imagine
21:22:32saratogawell i guess the DSP engine would have to handle figuring out when to resample
21:22:34saratogait could of course
21:22:58 Nick DormantBrain is now known as SuperBrainAK (~andy@shared02.balt01.cd.2g2u.net)
21:23:27saratogato be honest i haven't kept up with all the clean up its had over the last year or two
21:23:32saratogai need to look at it in more detail
21:24:07gevaertsIt could, but then the assumption of silence while switching doesn't hold any more
21:24:52saratogayou have to stop playback no matter what when changing the sampling rate on a lot of hardware, a PLL will have to relock or something similar
21:25:20gevaertsRight
21:26:01saratogai don't think theres any real technical or organizational reason we can't do this
21:26:07gevaertsSo that's at least a good reason not to force dynamic switching. I don't know if it needs a separate setting, or e.g. just depend on voice being used
21:26:17saratogajust seems more complex and error prone then just having a good resampler
21:26:34*gevaerts nods
21:26:35saratogawhich is the solution that virtually every other operating system uses
21:29:44saratogai think the only big problem our resampler has at the moment is that if theres ultrasonic signal, its probably only attenuated 20dB or so at most before its aliased
21:30:03saratogathat could be fixed with oversampling and low pass, but how common are strong 30kHz tones in real music?
21:30:40gevaertsNot very in any material that comes from a CD anyway
21:31:17saratogayeah, its mostly 96k material that might be a problem
21:31:27saratogawell if its stupidly mastered
21:34:54saratogawe could fix that by oversampling 2x (basically zero stuffing) and then using a low order LPF to filter out the ultrasonic stuff, then run that through the hermite resampler we have now
21:35:46saratogawould reduce aliasing by a factor of whatever the attenuation of the low pass was (probably 30 dB for a ~5MHz worth of CPU filter) so we could do a lot better
21:35:58saratogabut thats effort :)
21:36:54gevaertsAlso a bit of CPU, I imagine :)
21:37:11gevaertsAll for things that are arguably mastered wrong, even
21:38:03 Join kevku [0] (~kevku@2001:0:c38c:c38c:4e3:7336:3d69:be77)
21:39:45preglowsaratoga: why such bad rolloff?
21:50:18 Join AndChat|299081 [0] (~saratoga@2600:1008:b02c:79c8:0:42:3419:e201)
21:51:05***Saving seen data "./dancer.seen"
21:51:21AndChat|299081I think that's what the least squares solver in MATLAB gave me for a filter that was flat until 17 k and 10 the order at 96k
21:51:40pregloworder 10?
21:51:43AndChat|299081Ugh stupid chat lost my nick
21:52:08pregloworder 10 fir? that's nothing
21:52:24AndChat|299081That's why its fast
21:52:34preglowwell, yeah, but it doesn't do its job...
21:52:58preglowi'd rather try to fit in an iir filter at that complexity
21:53:03preglowif it fits in the algo
21:53:04AndChat|299081I think I got 30 db plus maybe 20 db from the polynomial
21:53:38AndChat|29908150 db rejection is probably good enough
21:53:52preglow50 db is nice enough in some circumstances
21:54:15preglowunluckilt, daps and headphones is a bit critical
21:54:28preglowwhich reminds me, i need to track down that goddamn mp3 decoder distortion issue on arm
21:54:36preglowit's starting to get to me
21:54:37 Quit AndChat|299081 (Client Quit)
21:56:55preglowalso, how does the resampler fare on 48k -> 44k?
21:57:03preglowprobably the most common case
21:57:55 Quit petur (Quit: Leaving)
21:58:23preglowoh well, beer time
22:00
22:16:01 Quit Zarggg (Quit: Zarggg)
22:23:32 Quit soap (Ping timeout: 248 seconds)
22:27:43 Join jhMikeS [0] (~jethead71@d192-24-174-117.try.wideopenwest.com)
22:27:43 Quit jhMikeS (Changing host)
22:27:43 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
22:28:37 Join soap [0] (~soap@rockbox/staff/soap)
22:29:58jhMikeSkugel: I know very little about the lang tools.
22:33:19jhMikeSperhaps I ought to just go ahead, though it will still break some compatibility if I merge the strings
22:46:41jhMikeSguess that was easy :)
22:48:02 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
22:54:28kugeljhMikeS: we don't usually make an effort to keep old lang files compatible
23:00
23:01:33jhMikeSkugel: but we do try to keep things compatbile when adding new strings or features. this is different in that I can reuse the string that is used for recording but it does potentially mess up indexes if no recording but there is play_frequency. Thinking about it further though, I'm not sure if any such targets exist that can do 48k but don't do recording, so really, the file wouldn't change in most cases.
23:01:39 Quit shamus (Read error: Connection reset by peer)
23:02:16 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
23:03:55 Join [ungali] [0] (~ungali]@24.244.29.57)
23:03:55 Quit [ungali] (Changing host)
23:03:55 Join [ungali] [0] (~ungali]@unaffiliated/ungali)
23:08:40 Quit [ungali] (Read error: Connection reset by peer)
23:10:42 Quit AlexP (Remote host closed the connection)
23:14:59 Join AlexP [0] (~alex@rockbox/staff/AlexP)
23:31:16 Quit belak (Ping timeout: 264 seconds)
23:33:53 Nick funman is now known as j-b (~fun@rockbox/developer/funman)
23:37:19 Nick j-b is now known as funm4n (~fun@rockbox/developer/funman)
23:38:05 Nick funm4n is now known as DGMurdockIV (~fun@rockbox/developer/funman)
23:38:29 Nick DGMurdockIV is now known as funman (~fun@rockbox/developer/funman)
23:43:18 Join belak [0] (~belak@facebook/engineering/belak)
23:50:43 Quit ladyblink (Ping timeout: 240 seconds)
23:51:07***Saving seen data "./dancer.seen"
23:52:03 Join ladyblink [0] (bassgeisha@selectah.drop.that.bass.aikyou.bassgeisha.com)
23:53:09*jhMikeS wants to put auto switch for playback under "no do"
23:56:35jhMikeS[Saint]: and no, crossfade won't be possible to do cleanly between files of different sample rates if trying to switch the hardware on the fly
23:58:34jhMikeSAnd it doesn't end there. There's voice mixing and delays in the DSP. I sooooo won't implement it because I can see impeding disaster and quite unpleasant side effects.

Previous day | Next day