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 2008-09-23

00:00:01 Join stripwax [0] (
00:00:04 Join matsl [0] (
00:00:14 Join shotofadds [0] (
00:00:15saratogai bet with proper scheduling good ASM for each platform, the Beast could probably have a 40-50% advantage
00:00:33saratoga(at equal clock speed)
00:01:22Lloreanamiconn: Agreed.
00:01:32gevaertsI agree. On the other hand, I think we can drop new bootloaders for the sansas (no real gain, and issues, so no sense spending time on it now)
00:01:37amiconnThis time there were loads of those
00:01:48*amiconn should probably make a set of flash images tomorrow
00:02:32gevaertsI'd say we release new bootloaders for all targets where we feel confident that they actually work, and maybe other targets later on
00:02:44amiconnDo we want some special version string for the new bootloaders?
00:02:54LloreanI think we should just increment their normal version numbers.
00:03:08LloreanSo v7 for the H120, v3 for the iPods (I think?) etc.
00:03:30amiconnsaratoga: I get 234% realtime on mini g2 for the same track. Does the vorbis codec differ in iram usage between PP5020 and PP5022?
00:03:41gevaertsIs rbutil ready to handle the release?
00:04:56amiconnThe iriver bootloaders still need to be tested by someone with access to a bdm, right?
00:05:57saratogaamiconn: sounds like it does
00:06:05saratogathough i'm not that familar with it
00:08:47 Quit saratoga ("CGI:IRC (EOF)")
00:10:00 Quit matsl (Remote closed the connection)
00:10:29 Quit Yondering (Read error: 110 (Connection timed out))
00:13:46 Quit bluebrother ("leaving")
00:14:04 Quit Schmogel (Read error: 54 (Connection reset by peer))
00:17:10 Join Schmogel [0] (
00:17:10 Quit mmadia (Read error: 104 (Connection reset by peer))
00:18:16 Quit bertrik ("Leaving")
00:18:47joetooI see that the Sansa e200 uses an FM chip with an upgrade path. It seems there is another chip that is pin compatible that adds RDS decoding. That's extra nice because I think RDS includes the accurate time.
00:19:42gevaertsjoetoo: thinking about hardware mods?
00:20:01 Join webguest87 [0] (n=485e7324@gateway/web/cgi-irc/
00:20:36joetoowhy not?
00:20:48webguest87how come my database on me sansa e260 isnt updated?
00:20:58joetoowebguest87: reboot yet?
00:21:01webguest87i put no songs in but it never shows
00:21:12 Quit mf0102 ("Ex-Chat")
00:21:20webguest87i rebooted and in database settings i click update now and still wnt work
00:21:31joetoono, update first, then reboot
00:21:39webguest87i did..
00:22:18stripwaxsorry, but why the need to reboot?
00:23:00joetooI have only had rockbox on my sansa for about an hour, but I found that updating the DB didn't seem to take until after a reboot.
00:23:36MarcGuaywebguest87: Are the songs tagged properly? Did you Initialize Now?
00:24:22gevaertswebguest87: does the database work for tracks that were on it earlier?
00:24:27webguest87tagged properly? what do you mean? and yeah i clicked intialized now but it says updating in backround
00:24:45webguest87and i needed to reboot before i use
00:25:34stripwaxwho says you need to reboot before you use it?
00:25:46gevaertsThe first time you do
00:25:59stripwaxoh, sorry, I see what you mean.
00:26:05webguest87it has to be tagged properly?
00:26:26webguest87the songs shows fine in Files>Music but not showing in Database
00:26:26joetoogevaerts: is that also true when swapping removable media?
00:26:28stripwaxthe media files need to have metadata tags (id3, etc)
00:26:47joetoostripwax: but they will at least show up as untagged
00:26:50stripwaxif the song is playing, you can see if the tags exist from the context menu
00:27:42gevaertsjoetoo: rebooting? no. It's just that rockbox doesn't allocate RAM for the database before "Initialize Now"
00:27:55webguest871 of my artist dnt show and new songs dnt show I need tags?
00:27:56joetoogevaerts: oic
00:28:08webguest87you are confusing me
00:28:26*gevaerts points everyone to the channel guidelines about using full words
00:28:43 Quit shotofadds (Read error: 104 (Connection reset by peer))
00:29:09joetoowebguest87: if the songs have no tags, I believe they will be listed as '<Untagged>' in each category.
00:29:37webguest87hold on let me try again
00:31:09webguest87nope still dnt work
00:31:28scorchewebguest87: stop saying "dnt" please...
00:31:50 Join fdinel [0] (
00:32:22webguest87i'm not talking to you and idc don't whatever...
00:32:32scorcheand "idc"...
00:32:34 Join mmadia [0] (
00:32:35gevaertswebguest87: please follow the rules
00:32:39Bagderwhat is idc anyway?
00:32:51scorchehave you looked at the guidelines for this channel?...they are linked to in the topic if you havent
00:33:30 Quit Nico_P (Remote closed the connection)
00:33:36webguest87my database isn't updating
00:33:49joetooBadger: I assume it is an acronym for "i [sic] don't care", but I could be wrong.
00:34:37 Join krazykit [0] (
00:43:32stripwaxwebguest87 - do your music files play correctly on rockbox (can you hear them), and do the tags show up correctly if you go to the context menu and view 'File Info'
00:46:43stripwaxSorry, "Show Track Info". I.e. the context menu described in section 4.3.3 of the manual
00:47:05 Quit herrwaldo ("Konversation terminated!")
00:47:16*stripwax notices a hanging sentence, "To access this screen," . hmmm.
00:47:49 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
00:48:31 Quit linuxstb (Read error: 110 (Connection timed out))
00:51:10joetooHow dishonest would it be to tell a chip manufacturer that your job title is "Senior Electrical Engineer" instead of "Senior, Electrical Engineering" when inquiring about ordering samples? ;-)
00:51:45 Join Zarggg [0] (
00:52:01*scorche moves joetoo over to #rockbox-community
00:52:05stripwaxwebguest87 - also, I think there is a menu option to view the database update status (assuming that Initialize Now did indeed start scanning your music and building the database).
00:56:13 Quit dabujo (Read error: 104 (Connection reset by peer))
00:57:29 Part pixelma
00:57:45 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
00:58:35joetooWhat does "Load to RAM" do for the database? Is that just to lower access time?
01:01:10scorchejoetoo: does the manual not describe that?
01:02:20*joetoo remembers the manual!
01:02:39*stripwax even linked to it, just a second ago
01:03:02 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
01:03:17*stripwax wonders if webguest87 is up and working now, or is just very silent
01:06:04joetoo"Note: If you browse your music should load to RAM...because the disk will not need to spin on each search." This line belongs in the Sansa e200 manual why again?
01:06:43gevaertsjoetoo: it's waiting for your patch
01:07:52dataangelWill rockbox puke if I have 5,000 files in a folder?
01:08:01LloreanYou'll need to change a setting.
01:08:04joetoo"Feature requests and documentation correction suggestions will only be accepted when accompanied by a patch." Eh?
01:08:30Lloreanjoetoo: One-word questions are hard to interpret...
01:08:34dataangelWill I need to change a setting just for the file viewer to not puke on that folder, or will I need it for rockbox even to run? The only thing that will access this folder is a plugin I'm writing, I'd never go to it by hand
01:08:47Lloreandataangel: File viewer.
01:08:53dataangelah k
01:09:20dataangelIs rockbox smart enough to only list max files in folder numbers that it can handle, or is it possible to pick one that exceeds available RAM?
01:09:42Lloreandataangel: It gives an error message if you go into a folder higher than the maximum you've told it to expect.
01:10:02LloreanThere are no negative effects beyond being unable to browse into that folder or properly playlist songs in it.
01:10:03dataangelright, but can I mistakenly tell it to expect a number that is actually greater than what the player can handle?
01:10:19dataangelah k
01:10:20dataangelthanks :)
01:10:47 Part Llorean
01:10:56 Quit tvelocity (Remote closed the connection)
01:11:04 Quit ender` (" Drive defensively. Buy a tank.")
01:11:15dataangelhaving my plugin have 5,000 individual files is easier than having it to try and load one gigantic text file and keep the whole thing in memory (and I need to insert text occasionally, so in one big file everything would have to be moved forward...)
01:12:26gevaertsdataangel: there is also a limit on the number of simultaneous open files
01:16:40dataangelgevaerts: that one I'm not worried about because I'll only ever have one open at a time
01:17:33 Quit Nibbler (Read error: 104 (Connection reset by peer))
01:19:17amiconndataangel: I don't know what kind of plugin you're working on, but keep in mind that having 5000 very small files will waste quite some disk space, depending on cluster size
01:20:19amiconnA fat(32) cluster can be up to 32KB (the compatible limit), so the potential diskspace requirement coult be 156MB, even if each file only contains a single byte
01:20:51amiconnAlso, going through a lot of small files (open -> read or write -> close) will take quite some time
01:21:25joetoodataangel: there is likely to be a fair amount of overhead associated with having so many small files. 5000 * the minimum filesize for a file system can be pretty large.
01:22:23*amiconn thinks there's an echo in here...
01:22:33 Join Rondom [0] (
01:22:38***Saving seen data "./dancer.seen"
01:23:10stripwaxme sighs
01:23:20 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
01:23:21*gevaerts hands stripwax a /
01:24:03*joetoo didn't notice because he was researching FAT32 at the time.
01:25:55 Quit Hillshum (Read error: 110 (Connection timed out))
01:32:41joetooI am now waiting for SANYO to send me information about a chip with which to upgrade my Sansa e200's FM tuner to support RDS. Here's crossing my fingers that small quantities and/or [free] samples are available.
01:33:45 Quit coatman (Remote closed the connection)
01:35:21dataangelFinding out what files are in a folder using readdir() (in the plugin API) doesn't count those files as open towards the max open file limit, does it?
01:37:22 Join coatman [0] (
01:37:45 Quit MethoS-- (Remote closed the connection)
01:38:29 Join kushal_12_27_200 [0] (n=kushal@
01:42:36 Quit perrikwp (" ajax IRC Client")
01:47:42 Join ajonat [0] (n=ajonat@
01:55:15 Nick HBK- is now known as HBK (
02:05:48 Join ladeh_k [0] (n=76d04a15@gateway/web/cgi-irc/
02:07:03*amiconn is working on different controls for cube.rock
02:07:21amiconnThe current ones contain some impossible combos on several targets
02:07:42amiconnI wonder whether this would be 3.0 stuff (cosmetic, but...)
02:08:46joetooI can't find the test_codec in the plugins menu. Does it hook into the record menu directly? Or does it get excluded while building?
02:09:13scorchejoetoo: you need to enable it in the SOURCES file
02:09:20joetooscorche: ah, thanks.
02:09:35amiconnjoetoo: ..and then you still won't find it in the plugins menu, as it's a viewer
02:09:41amiconnUse "Open with..."
02:25:47 Join DaCapn [0] (
02:27:31 Quit jon-kha (Read error: 145 (Connection timed out))
02:35:16joetooI see the pitch shift option, but I'm surprised that I don't immediately find the option to change the playback rate while maintaining the original pitch. Am I looking in the wrong places?
02:37:58MarcGuayjoetoo: I don't believe it's been put into the regular builds yet.
02:47:43joetoooh ok. thanks. I'll look for it there and in svn.
02:48:00 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
02:49:45saratogaamiconn: AC3 is actually real time for me using the sample I found online
02:49:46 Join mmadia_ [0] (
02:49:51saratogagot a particular sample I should try?
02:50:42amiconnWell, my samples weren't realtime on PP, not even 2.0, let alone 5.1
02:51:15amiconnOn coldfire they are realtime. 2.0 is very fast there
02:51:18saratogawell i'm about to hopefullly change that
02:51:38saratoga38% speed up for about 8 lines copied and pasted out of libwma
02:51:52amiconnMy samples are all 48kHz though (audio tracks ripped from dvd)
02:52:32saratogathis codec is so simple, could probably get my random sample over 300% realtime just stealing bits from the other transform codecs
02:53:31 Quit coatman ("Leaving")
02:53:39 Join coatman [0] (
02:55:26amiconnDoes it use emac assembly on coldfire?
02:55:57saratogajust for the fixed mul functions
02:56:01saratogaeverything else is plain c
02:56:32amiconnHmm, interesting. Probably plenty of iram usage then (iirc the codec is quite small)
02:56:39mmadia_soap ?
02:56:49saratogayeah someone went through and IRAMed all the tables
02:56:58soapmmadia, ?
02:57:49saratogathe IMDCT could be switched over to the rockbox one, but its annoying to do since the format stores some of the fourier coefficients interleaved under the expectation that you'll use the FFT based IMDCT like WMA used to use before the tremor stuff was added
02:57:57mmadia_soap : as i've been going through some e200 themes, would you want me to list the themes that are known working and which ones i've fixed (for e200 only)
02:58:54saratogaeveryone seems to like that form of the IMDCT better
03:00:05mmadia_soap : about 40 fixed themes are waiting to be uploaded ( and i still need to request wiki write access)
03:00:35soapmmadia, yes. Start a new thread (something like) "Wiki Theme Status" and post yes/no for all the themes you have tested. I'll merge the posts into one master list.
03:01:00amiconnsaratoga: Iirc someone (preglow?) said in here that implementing the imdct using fft and some additional processing would be faster (?)
03:01:24 Quit culture (Connection timed out)
03:01:25saratogaamiconn: yeah thats been my theory, but i've not found an FFT implementation on ARM where that is the case
03:01:48amiconnAnd on coldfire?
03:01:52 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
03:02:13saratogaI don't have anyway to test, but so far the Tremor way has been much faster then either of the FFT ways in Rockbox
03:02:44saratogaits somewhat surprising that no fast open source ARM FFT exists
03:03:24saratogai'm less surprised that no such FFT has been found for coldfire
03:03:26amiconnI am wondering about something that's transform related: While coldfire is fast at decoding mp3, and arm being slower, it's the opposite for mpeg2 video decoding. Do you have an idea why this is so?
03:04:36saratogathe transforms are nearly identical between the codecs [mpeg2 uses a DCT, and libmad implements a lot of the synth filter with a DCT], so I'd guess maybe mpeg2 uses a lot of DRAM
03:04:48saratogabut its just a guess
03:04:59saratogai suppose you could profile mpegplayer and see where it spends all its time
03:06:05saratogai know ARM is slow in mp3 because the synth filter does an enormous number of "load word then fix point multiply it and then store it" loops
03:06:22amiconnThis is C or asm?
03:06:39amiconnHmm. Does the asm use ldm?
03:06:42 Quit Rondom ("Ex-Chat")
03:06:52saratogasome, but half the loads are spaced at 8 byte intervals
03:07:08saratogayeah no fun
03:07:27saratogatheres probably some way to rewrite teh algorithm to make it more friendly on ARM
03:07:29dataangelCan I use double's in rockbox plugins? I know they'd be slow, but can I use 'em?
03:07:43saratogabut I don't understand it well enough to do that, so I just spun it off onto the COP
03:07:53saratogawhich happily enough worked pretty well
03:08:11amiconnThe mpeg2 idct allows some trickery. if you store the intermediate results in transposed form, you can use ldm (or movem on coldfire) for the 2nd part as well
03:08:13 Join pabs_ [0] (
03:08:46saratogathats certainly nice
03:08:53amiconnI know the imdct for audio isn't 2-dimensional, but perhaps some similar tricks could be applied?
03:09:18saratogafor mp3 or for the other codecs?
03:09:54amiconnEither of them
03:10:38 Join AndyIL [0] (i=AndyI@
03:11:02saratogai haven't looked at the asm for the Tremor IMDCT actually
03:11:03saratogai should
03:11:30saratogalooking at the MAD IMDCT i wonder if it could be done a bit better in assembly
03:11:53 Quit jhMikeS (Nick collision from services.)
03:11:59 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
03:13:23saratogait does seem to do a lot of ldr's in the windowing code, which is almost certainly slower then LDM
03:13:57saratogathough my ARM asm is rusty
03:13:58amiconnUnfortunately the mpeg2 idct cannot make use of ldm, because the data is 16 bit
03:15:12amiconn...and it's signed. Splitting 32 bits needs 3 additional instructions. It would actually be marginally faster on arm7, but slower on any higher arm
03:15:52 Quit coatman (Remote closed the connection)
03:17:20saratogaamiconn: did you try that latest commit on your AC3 file?
03:19:17 Join coatman [0] (
03:19:21amiconneh, not yet
03:19:44saratogaif its not real time yet i can look into it
03:20:22saratogaanyway i'm going home
03:21:49 Quit AndyI (Read error: 110 (Connection timed out))
03:22:41***Saving seen data "./dancer.seen"
03:22:59 Quit coatman (Client Quit)
03:23:04 Join coatman [0] (
03:23:48 Quit saratoga ("CGI:IRC (EOF)")
03:23:58 Quit webguest87 ("CGI:IRC")
03:24:27 Quit pabs (Read error: 110 (Connection timed out))
03:27:28 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
03:29:19 Quit MarcGuay ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
03:30:16mmadia_soap : topic created.
03:35:46soapmmadia_, thank you. I'll try to organize this, but real life has firm hold of me for at least the next 48 hours.
03:40:28 Quit massiveH ("Leaving")
03:43:32mmadia_it'll give me time to add to the list ; )
03:44:54 Quit krazykit ("laters.")
03:47:24 Join Llorean [0] (n=DarkkOne@
03:52:16 Quit DerDome (Nick collision from services.)
03:52:17 Join DerDome1 [0] (
03:52:27 Nick DerDome1 is now known as DerDome (
04:00:00 Quit mmadia_ ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
04:07:31 Join miepchen^schlaf_ [0] (
04:09:32 Quit Lambduh (Read error: 104 (Connection reset by peer))
04:09:35 Join Lambduh [0] (
04:19:41 Part mmadia
04:21:19 Join Davide-NYC [0] (
04:22:05 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:22:11 Join massiveH [0] (
04:22:17Davide-NYCI saw a change to test_codec specific to the GigabeatS.. is it appropriate to run codec tests and post the results in the wiki now?
04:22:57LloreanI wouldn't suggest adding them to the wiki until it's a supported target.
04:23:42 Join m0f0x [0] (
04:24:16 Join JdGordon|uni [0] (i=768aa358@gateway/web/ajax/
04:27:11Davide-NYCLlorean: will hold off then
04:27:21 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
04:29:12 Quit Davide-NYC ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
04:33:44 Nick fxb is now known as fxb__ (
04:35:02 Quit ajonat ()
04:35:57 Quit JdGordon|uni (" ajax IRC Client")
04:44:51 Join jeffdameth1 [0] (
04:59:00 Quit jeffdameth (Connection timed out)
05:07:39jhMikeSsaratoga: (logs again) no, the patches don't deadlock. the key is that "PCM ready" is equivalent to "synth idle".
05:10:02 Quit Seed ("cu, Andre")
05:11:33 Join JdGordon|uni [0] (i=768ae223@gateway/web/ajax/
05:16:19jhMikeSamiconn: (for logs) Should singe-core builds be still threaded the same way but run only on CPU or should single core builds be exacly like a single-core target?
05:17:06jhMikeSthough the "core" parameter to create_thread doesn't exist in that case
05:22:43***Saving seen data "./dancer.seen"
05:25:10 Quit kushal_12_27_200 (Connection timed out)
05:26:56 Quit massiveH ("Leaving")
05:30:38 Quit XavierGr ()
05:32:01 Join amdgoon [0] (n=amdgoon@
05:34:04 Quit JdGordon|uni (" ajax IRC Client")
05:35:55 Quit Zarggg ()
05:36:23 Quit m0f0x ()
05:37:19 Join Zarggg [0] (
05:43:18 Quit Zarggg ()
05:45:51 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
05:48:44 Quit Horscht ("User was distributing pornography on server; system seized by FBI")
05:49:43 Join Zarggg [0] (
05:50:56 Quit dataangel ("Leaving")
06:00:16 Quit homielowe_ ()
06:04:29 Join JdGordon|uni [0] (i=768ae23b@gateway/web/ajax/
06:06:55 Join homielowe [0] (
06:11:45 Join reacocard [0] (n=reacocar@WL-271.CINE.HMC.Edu)
06:14:42 Quit jhulst (Remote closed the connection)
06:27:03 Quit Zarggg ()
06:27:56 Join Darksair [0] (n=user@
06:28:42 Quit daurnimator (Read error: 60 (Operation timed out))
06:28:49 Join daurnimator [0] (i=daurn@unaffiliated/daurnimator)
06:31:16 Quit EspeonEefi ("さよなら")
06:31:59 Quit Darksair (Remote closed the connection)
06:32:09 Join Darksair [0] (n=user@
06:37:12 Quit JdGordon|uni (" ajax IRC Client")
06:41:01 Quit Siku ()
06:48:39 Quit Darksair ("ERC Version 5.3 (IRC client for Emacs)")
06:52:10 Join Darksair [0] (n=user@
07:01:22 Quit ladeh_k ("CGI:IRC (EOF)")
07:12:55 Join ladeh_k [0] (n=76d04a15@gateway/web/cgi-irc/
07:12:55 Quit ladeh_k (Client Quit)
07:22:09 Join ladeh_k [0] (n=76d04a15@gateway/web/cgi-irc/
07:22:47***Saving seen data "./dancer.seen"
07:23:19 Quit Llorean ("Leaving.")
07:23:26 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
07:25:06 Quit perrikwp (" ajax IRC Client")
07:26:58 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
07:36:27 Quit ladeh_k ("CGI:IRC (EOF)")
07:36:47 Quit cool_walking_ (" ajax IRC Client")
08:01:37 Join einhirn [0] (
08:08:50 Quit miepchen^schlaf_ ()
08:26:13 Quit faemir (Remote closed the connection)
08:27:58 Quit BHSPitMonkey (Remote closed the connection)
08:34:11 Quit BigBambi (Read error: 113 (No route to host))
08:34:23 Join bluebird [0] (i=rbeindeo@
08:39:05 Join JdGordon|uni [0] (i=795b5418@gateway/web/ajax/
08:43:24 Nick JdGordon|uni is now known as JdGordon| (i=795b5418@rockbox/developer/JdGordon)
08:47:52amiconnjhMikeS: They should behave like single core targets
08:49:57 Join ladeh_k [0] (n=76d04a15@gateway/web/cgi-irc/
08:50:27 Join ender` [0] (
08:52:04 Join Rob2222 [0] (
08:52:07amiconnjhMikeS: Sometimes it is useful to be able to go back to a less functional state of some parts of rockbox, e.g. it was very useful to be able to compile an Ondio build without multivolume and hotswap support in order to debug the weird Transcend MMC behaviour
08:57:27 Quit JdGordon| (" ajax IRC Client")
09:04:34 Join petur [50] (n=petur@rockbox/developer/petur)
09:10:11 Quit Rob2223 (Read error: 110 (Connection timed out))
09:13:14 Quit ladeh_k ("CGI:IRC (EOF)")
09:14:12 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
09:14:42 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
09:15:04 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
09:15:07 Join BigBambi [0] (i=86ceaf40@rockbox/staff/BigBambi)
09:15:30amiconnsaratoga (logs): I'm quite puzzled about my a52 speed tests
09:16:21amiconnI cannot measure a difference with r18570 on PP502x
09:17:33amiconnBut then this sample is also realtime by a nice margin. I know I had some which weren't, but I can't find them atm
09:22:49***Saving seen data "./dancer.seen"
09:25:50amiconnAh, 'make' dependency artifact :\
09:27:47 Join reacocard_ [0] (n=reacocar@
09:31:07amiconnNow I see the speedup (192->307%, even more than r18570 claims), but I still wonder about my non-realtime samples....
09:40:03 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
09:41:38 Quit reacocard (Read error: 110 (Connection timed out))
10:00:13 Join advcomp2019_ [0] (
10:00:13 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
10:00:20 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
10:00:54 Quit bughunter2 ("bye")
10:06:00 Join lasser [0] (
10:07:41GodEaterit's release day today isn't it ?
10:10:02jhMikeSwe're done? that's it? we can all go home now? :p
10:11:35GodEaterturn off the lights when you leave
10:20:06 Join culture [0] (
10:21:33petur(if you're the last one to leave)
10:24:27 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
10:35:05 Quit vitja (Remote closed the connection)
10:35:22 Join vitja [0] (n=vitja@
10:35:48vitjagevaerts: did you have some success?
10:42:48gevaertsvitja: I got it running about as well as you I guess. I'll work more on it tonight
10:43:27vitjagevaerts: does you host detect it as flash?
10:45:34gevaertsvitja: not yet no. I don't think I had actual bulk transfers working. No time to look now though
10:49:15 Join Nibbler [0] (
11:03:14 Join AndyI [0] (n=pasha_in@
11:11:10amiconnjhMikeS: There are coding style violations in mpa.c. While the general brace placement style is selectable, the opening brace of a function is *reqired* to be placed in the next line
11:14:08 Quit AndyIL (Read error: 110 (Connection timed out))
11:17:45jhMikeSamiconn: I supposed I would do style changes separately in a code policing commit if I did it
11:20:47jhMikeSmight as well do so now, what the heck
11:21:28 Join tvelocity [0] (n=tony@
11:22:52***Saving seen data "./dancer.seen"
11:52:56JdGordonwho was complainig about the #ifdef's in plugin.[ch] not being the correct ones around the added kernel stuff?
11:55:26JdGordonjhMikeS: all swcodec has semaphore and even objects right?
11:55:39JdGordonor should i bump the plugin api to fix the #ifdefs?
11:57:19JdGordon... I guess they do or we wouldnt have a green table
11:57:27jhMikeSJdGordon: who was compaining? I don't know. But yes, swcodec does but any could.
11:57:54JdGordonI dont remember who it was
11:58:02jhMikeSin kernel.c or thread.c?
11:58:04*JdGordon double checking the patch compiles and will comit
11:58:12jhMikeSoops, right
11:59:20 Join MethoS- [0] (
11:59:30jhMikeSit is broader than current useage, but not technically incorrect I guess
12:00:54JdGordonuntil someone decides to compile swcodec without either of them
12:01:11JdGordonbut yes, atm its just a better safeguard or something
12:02:40 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:03:00jhMikeSactually, no code in single core uses those either (anywhere). they're sort of just taking up space there for the moment.
12:03:16linuxstbjhMikeS: I saw that the event and semaphore functions themselves were surrounded by something like #ifdef HAVE_SEMAPHORE_OBJECTS, but the use of those functions in plugin.[ch] used SWCODEC (and your older commit to codec.[ch] I think). It seems inconsistent.
12:04:28 Quit culture (Read error: 104 (Connection reset by peer))
12:04:45jhMikeSlinuxstb: SWCODEC is what defines those certain objects to be included, but for codecs only NUM_CORES > 1 was actually using it
12:04:46JdGordonah, its NUM_CORES > 1 in the codec api
12:05:54JdGordonimo it should be consistant with the actual function, not whats using it (for firmware/ level functions only though, apps is another story)
12:08:32linuxstbjhMikeS: Those objects are unused on single-core SWCODEC targets, but are still compiled?
12:10:07jhMikeSlixuxstb: yes, config.h could be refined to keep it out until used.
12:10:26jhMikeSthe event is used, semaphore, no
12:12:45jhMikeSsemaphore _was_ used when we had voice swapping
12:23:08 Quit bluebird (Read error: 54 (Connection reset by peer))
12:28:22 Join culture [0] (
12:30:33JdGordonis 32 chars enough for most of the possible id3 fields?
12:30:49JdGordonim tihnking band and track title might be the only ones which would come close?
12:32:04jhMikeS32? not with my stuff
12:32:26JdGordonnot enough? or not close to being a problem?
12:32:52jhMikeSalot of classical stuff would go way over that
12:34:07JdGordonlike what?
12:34:19JdGordon31 letters is a pretty decent amount...
12:34:25GodEaterno it's not
12:34:29jhMikeSbo beep - mary had a little lamb <= 33 letters
12:34:31markunJdGordon: who are you to decide that?
12:34:35GodEaterClassical stuff wouldn't even come close to fitting in there
12:35:06JdGordonmarkun: I'm not... just trying to implement 8955 a bit nicer
12:36:09markunJdGordon: even after reading that page I have no idea what it does :)
12:36:53JdGordonit removes id3 fields which dont have any data so the list might actually fit on one screen
12:36:57JdGordonor be more readable
12:37:17JdGordonI wanted to use the simplelist buffer, but its currently only got room for 32 32char strings
12:40:19 Join dany_21a [0] (
12:45:09JdGordonok then... back to finshing off the bloody bookmarks patch :/
12:52:45 Join dabujo [0] (
12:54:36 Join BlakeJohnson86 [0] (
12:59:20JdGordonwould anyone object to me adding a line in the rockbox info screen saying the AA size the current wps expects?
13:02:16 Join cool_walking_ [0] (
13:10:08cool_walking_I'm trying to "unbrick" a Sansa e260 (following SansaE200Unbrick), but none of the Recovery, Manufacturing, or Pre-boot modes seem to work. I don't get any USB connectivity or indication that I've entered that mode (screen stays completely black). The only feedback I have is the lighting up of the wheel, and the fact that (most of the time) it turns off when holding "power/menu" for 30 secs.
13:12:55*JdGordon doesnt really see the need to speak the line saying the AA dimensions...
13:12:57amiconnJdGordon: 32 chars would be way too short for some of my tracks
13:13:24JdGordonok ok... I've moved on from that :p
13:16:48JdGordonalso... why is disk free space above disk size in the info screen?
13:16:56*JdGordon tihnks they shold be swapped
13:20:27*JdGordon sits back and waits for the tongue lashing!
13:22:54***Saving seen data "./dancer.seen"
13:23:05 Join MarcGuay [0] (
13:25:16JdGordonis the number order in configure really important?
13:25:40JdGordonI think Philips and Meizu should be swapped to maybe make that last row easier to read?
13:25:49JdGordonor can we add some extra spaces in?
13:28:25JdGordonas long as we stay under 80chars wide its OK yeah?
13:28:39JdGordonor are consoles expected to be narrower?
13:29:27 Join LambdaCalculus37 [0] (
13:30:09JdGordonBagder: 70chars at its widest is still OK yeah?
13:30:27Bagder78 or so should be fine
13:30:34JdGordonok cool
13:40:01 Quit tvelocity (Read error: 110 (Connection timed out))
13:40:42 Join advcomp2019__ [0] (n=advcomp2@unaffiliated/advcomp2019)
13:40:47 Quit jhMikeS (Nick collision from services.)
13:40:53 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
13:40:53 Quit advcomp2019_ (Read error: 104 (Connection reset by peer))
13:44:30 Join Dementio [0] (n=dementio@
13:47:42Dementioanyone know how to compile in a default hard drive spin down time?
13:48:16Dementioto be more specific, i need to turn it off completely
13:48:37 Nick reacocard_ is now known as reacocard (n=reacocar@
13:53:40 Quit LambdaCalculus37 ("Do quit now, there's a demon around the corner!")
13:57:03 Join J-23 [0] (
14:03:16JdGordonDementio: why would you want to?
14:03:55JdGordonthe spin down time is only used if the user made the disk spin.. if its just rebuffering it turns off automatically as soon as its done (iirc)
14:06:15 Join LambdaCalculus37 [0] (i=44a04303@gateway/web/ajax/
14:06:23Dementioi have an ipod video with the cfmod, i finally got it to boot by changing two lines in system-99502xx.c, but now it hard locks once the spindown time is reached
14:06:50Dementioas in once it boots, i had less than 4 seconds to change the setting from 4 seconds
14:07:16Dementioi was finally able to get it set to 254, but it still locks once that time is reached
14:07:38*JdGordon thought the cf mod worked fine on all targets now
14:07:51Dementionot on all cf cards
14:08:01Dementiotrying to get it working with a-data's
14:08:16Dementiowhich seems to be the one most people bought
14:08:38JdGordonamiconn: didnt you fix it so cf cards worked correctly?
14:08:44Dementioit still gives an ata error: -1 with the default code
14:11:30DementioPS: this was with r18563M-080922 from SVN
14:14:14 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
14:16:38 Part dany_21a
14:21:23MarcGuayDoes BCB mean anything to anyone? Boot Code Block? Boot Code Binary?
14:23:56 Join dany_21a [0] (
14:25:23markunMarcGuay: what's the context?
14:26:49 Quit courtc_ (
14:26:49 Quit Zom (
14:26:49 Quit Haudrauf (
14:26:49 Quit fxb__ (
14:26:49 Quit tecnmmag (
14:27:02MarcGuayThe Zen V Plus firmware doesn't have a FBOOT file the way the ZVM does. Instead it has a very small FBCB file which contains the strings STMP (the chip), BCB2, and CACA. I'm thinking it's just filler saying "the bootloader is on the ROM in the STMP".
14:27:31NJoinfxb__ [0] (
14:27:47 Join courtc [0] (
14:27:48MarcGuayAm curious if BCB is a known acronym.
14:28:06NJoinZom [0] (
14:28:17NJoincourtc_ [0] (
14:28:17NJoinHaudrauf [0] (
14:28:17NJointecnmmag [0] (n=tecnmmag@
14:29:28 Quit courtc_ (Connection reset by peer)
14:29:29 Quit Dementio (Read error: 54 (Connection reset by peer))
14:29:35LambdaCalculus37MarcGuay: Did you try punching it into Wikipedia or Google?
14:29:38 Join Dementio [0] (n=dementio@
14:30:09 Quit Haudrauf (Excess Flood)
14:30:23 Join Haudrauf [0] (
14:31:07MarcGuayLambdaCalculus37: Absolutely. I wouldn't ask before researching first.
14:31:23GodEaterMarcGuay is a good little netizen ;)
14:31:41MarcGuayCivic pride
14:31:45LambdaCalculus37Ha! Wikipedia gave me "Borland C++ Builder"! :P
14:32:35Dementiook, can i possibly fry my ipod if i keep changing certain parts of ata.c?
14:32:52 Part TMM ("Ex-Chat")
14:33:12GodEaterDementio: I can't think of a way you can
14:33:26Dementiook, good
14:33:41 Join Lambdumb [0] (
14:33:57DementioBuffer Control Block?
14:34:02 Quit Lambduh (Connection reset by peer)
14:35:18Dementioanyway, what if i just made it reference ata_flash.c instead of ata.c?
14:35:23Dementioi really don't know
14:35:35Dementionot much of a programmer
14:35:53Dementioi know
14:36:10DementioMarcGuay was asking what BCB stood for
14:36:28LambdaCalculus37Dementio: Except that Buffer Control Block is on that list.
14:36:42LambdaCalculus37MarcGuay: I also got Block Control Byte.
14:37:58 Join {phoenix} [0] (
14:38:10JdGordondoes the ATA spec have a no-op command?
14:38:26GodEaternot that I recall
14:39:02JdGordonDementio: I was hoping amiconn would help you out... anyway, you can comment out the block in firmware/drivers/ata.c in ata_perform_sleep()
14:39:20MarcGuayLambdaCalculus37: I guess I didn't look as hard as I thought, thanks.
14:39:24Dementioi'll try that now
14:39:55JdGordonGodEater: yeah, I didnt think so but changing 1 value is easier than finding the right code to comment out for him :)
14:40:01LambdaCalculus37MarcGuay: No worries. :)
14:40:24GodEaterJdGordon: you are the epitomy of the lazy programmer ;)
14:40:49JdGordonno! it seems there is a NOP command
14:40:59JdGordonand since when is being lazy bad for a progreammer? :D
14:41:41 Part LinusN
14:42:53GodEaterI never said it was *bad* :)
14:42:53amiconnThe proper thing to do would be to find out which command hangs, and in what way, and then fix the driver to handle this
14:44:02GodEaterJdGordon: I seem to recall you getting punched once as a result :) I even have the video somewhere!
14:44:18JdGordonthat wasnt me being lazy...
14:44:32JdGordonthat was not testing enough, or something, or red delta...
14:45:17GodEaternot testing could be considered lazy
14:45:29GodEaterI don't think it was a red delta though
14:45:51JdGordonamiconn: shouldnt we be checking the disk actually supports sleeping before calling SET_REG(ATA_COMMAND, CMD_SLEEP);?
14:46:54JdGordonor is that done elsewhere?
14:47:11amiconnThe problem is that many CF's report that they support it, but they don't -> spec violation
14:47:53JdGordonso apart from custom builds, there really isnt much we can do?
14:47:55Dementioamiconn, what is seems, is that anything that tries to write to the drive after the spindown time has passed
14:48:03Dementiosorry, was on console
14:54:04DementioJdGordon, which ata_perform_sleep() do i comment out? there's two
14:54:50JdGordonor just comment out the code inside that function
14:57:12Dementio<−−−−first time with c
14:57:48Dementioi was just gonna delete the line
14:59:46 Quit MarcGuay ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
15:03:30 Quit shodanX (Remote closed the connection)
15:08:17 Part vitja ("Ex-Chat")
15:09:37Dementionow if only my ipod would get enough of a charge to run
15:12:48peturhmmm I thought all CF cards were supported...
15:14:14peturand this is with a bootloader and build from current svn?
15:17:43peturand it boots and runs right up until the spindown?
15:18:13Dementioworks fine till then, once "DEV_RS = 0x3ffffef8;" is commented out
15:18:27Dementiootherwise i still get an ATA Error: -1
15:18:59LambdaCalculus37What kind of CF card is it?
15:19:06Dementioa-data 32gb
15:21:10peturbah.. I have an A-Data 8GB and Transcend 32GB...and no ipod to test in anyway
15:22:53peturpart of the fix for CF was to always do an init, even after a soft reset (which should keep the config), so we had a similar issue fixed :(
15:22:55***Saving seen data "./dancer.seen"
15:23:53 Join moos [0] (
15:24:14Dementioi tried manually changing the setting in the config file while in disk mode, but it doesn't use it. is it suppose to be that way?
15:24:44Dementioit would only use the setting if it was selected in the menu
15:24:49peturI meant the CF configuration that is done at startup (pio mode,...)
15:25:06 Join BlakeJohnson861 [0] (
15:25:43petureuh... you mean rockbox doesn't load its config file?
15:26:14Dementionot if i type in the settings manually
15:26:42 Join Schmogel [0] (
15:26:52Dementioin the config.cfg file
15:27:08peturbut any settings you save are loaded again?
15:27:25peturI suspect you type them wrong
15:27:43Dementioi copied and pasted out of another config
15:27:49peturor some CR/LF or encoding issue?
15:27:53Dementiodisk spindown: 254
15:27:56Dementioi use vi
15:28:07peturI've edited them lots of time
15:30:03Dementioi just got a completely fresh svn copy, and i'll only comment out "DEV_RS = 0x3ffffef8" and try to load the config
15:30:13 Quit linuxstb (Read error: 110 (Connection timed out))
15:30:26 Join shodanX [0] (
15:31:32 Quit J-23 (Remote closed the connection)
15:34:18 Join J-23 [0] (
15:35:16 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
15:37:39 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
15:40:37Dementiowhat happens if i set disk spindown to 0?
15:41:05 Join lazka [0] (
15:41:43lazkarecent build crashes on my sansa right after the intro screen
15:41:53JdGordonhow recent?
15:41:57JdGordonwhich sansa?
15:42:19JdGordondid you extract the whole zip file or just rockbox.mi4?
15:42:25*markun blames JdGordon :)
15:42:33 Part nicktastic
15:42:51lazkai'll try the zip..
15:50:40lazkaok.. zip works..
15:51:26Dementiook, not even with changing DEV_RS, i still get an ATA Error: -1
15:54:10 Part lazka ("cya")
15:54:56 Join MethoS-- [0] (
15:55:54 Quit JdGordon ("Konversation terminated!")
15:58:59Dementiothanks for the help guys, i'm quitting for tonight
15:59:22 Join JdGordon [0] (
16:04:25 Join mf0102 [0] (
16:05:29 Join n1s [0] (n=nils@rockbox/developer/n1s)
16:14:55 Quit MethoS- (Read error: 110 (Connection timed out))
16:19:21 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
16:24:41 Nick pabs_ is now known as pabs (
16:25:51 Quit {phoenix} (Remote closed the connection)
16:26:36 Quit DaCapn ()
16:31:35 Join Hillshum [0] (
16:32:26 Join goffa_ [0] (n=goffa@
16:37:23 Quit pixelma2 (Nick collision from services.)
16:37:26 Join pixelma2_ [0] (
16:37:36 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
16:37:41 Quit amiconn (Nick collision from services.)
16:37:48 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
16:42:08 Join toffe82 [0] (
16:44:33 Join jeffdameth [0] (
16:45:32 Quit goffa (Read error: 110 (Connection timed out))
16:59:31 Quit jeffdameth1 (No route to host)
17:01:42 Join jgarvey [0] (
17:11:29 Part lasser
17:15:16 Join desowin [0] (
17:18:40 Join bughunter2 [0] (n=Jelle@
17:20:14 Join pixelma [0] (i=d5494da5@gateway/web/ajax/
17:21:16*pixelma wonders why JdGordon had to invent a new feature for album art related language strings - is lcd_non-mono not good enough?
17:22:59***Saving seen data "./dancer.seen"
17:24:00 Join Jungti1234 [0] (n=742a2d5b@gateway/web/cgi-irc/
17:28:01pixelmalooks like some of the ports in progress but already in the buildtable don't have album art even though they don't have monochrome displays (M:Robe500 + the Zens)
17:29:19 Quit Jungti1234 ("CGI:IRC (EOF)")
17:29:19 Quit Dementio (Read error: 104 (Connection reset by peer))
17:29:30 Join Dementio [0] (n=dementio@
17:29:53pixelmaand I'm not sure I ever would expect the album art size stated in the Rockbox Info screen...
17:30:47LambdaCalculus37pixelma: I was wondering how that's supposed to work (the album art size info).
17:30:55pixelmathat too
17:31:26LambdaCalculus37Does it show you the album art size according to the cabbiev2 theme? Or your current WPS? I'm puzzled.
17:31:43*LambdaCalculus37 thinks we should pester JdGordon|zzz later, when he's awake
17:35:48BigBambiCurrent WPS according to what he said earlier
17:36:28LambdaCalculus37BigBambi: Ahh, okay. Thanks!
17:37:25pixelmaunrelated, but I also wonder whether I should try to check in my few changes to the manual (mostly cosmetic ones) from here now. I think they should go into the release branch too... not very comfortable to do here but I'm not sure how much time I'd have to do so later but before the release :\
17:39:16pixelma"few" as in "a number of files but no major changes... except the text_viewer one
17:39:20LambdaCalculus37pixelma: If it's mostly cosmetic, I say go for it.
17:39:50 Join Hillshum_ [0] (n=4ba5e7c0@gateway/web/cgi-irc/
17:41:18 Join Llorean [0] (n=DarkkOne@
17:41:36Hillshum_query danny_21a
17:41:43 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:42:11LambdaCalculus37Hillshum_: One less n. ;)
17:42:24Hillshum_query dany_21a
17:42:35pixelmaand a / at the start
17:44:04dany_21ahello again
17:44:54Hillshum_had to push the video card in, try to boot off old disk with ubuntu, and wait for live cd to boot
17:45:45Hillshum_so i'm not on my system i am devleoping on
17:46:19dany_21ayou asked for support while patching a non E200-Player - or?
17:46:59 Quit cool_walking_ (Remote closed the connection)
17:47:11Hillshum_i want to help port e200v2, but need to learn how to doo it...
17:47:26 Join cool_walking_ [0] (
17:48:15dany_21aah... so you have a e200? thats good - bec. on that device we can gain direct access to the flash over USB, so you can easily unbrick it
17:48:24dany_21abut you have to open the player for that
17:48:38*Hillshum_ needs to clean it anyway
17:48:54dany_21aokay :)
17:50:12dany_21asearch for that open solder pads (the two on the top of that pic)
17:50:33dany_21aand solder two wires two them and connect them via switch
17:51:01Hillshum_i've taken it fully apart many a time
17:51:36dany_21a(you dont need to solder a switch to them - you can also bridge them on demand with a piece of wire... but its much more comfortable)
17:52:08*Hillshum_ would like to avoid soldering
17:52:45 Quit pixelma (" ajax IRC Client")
17:52:50dany_21aanyhow... just try once to get into the recovery mode, only to check if it works, when needed :)
17:53:28dany_21abtw. i also use that mode to upload a new firmware - so i have only to upload some bytes, and not the whole firmware all the time - saves a lot of time
17:55:22Hillshum_leave the bat out?
17:56:05dany_21ayou can power it via usb - i have mine connected to a external power supply
17:56:05 Quit Llorean (Read error: 104 (Connection reset by peer))
17:58:09Hillshum_the inner screws are not cooperating
17:58:20 Quit BigBambi (" ajax IRC Client")
17:58:35Hillshum_the hard to take out ones
18:02:29Hillshum_my mouse is on one usb port and my dsl modem on the other
18:06:18dany_21ayou mean you have no free usb-port? take the mouse away, temporary
18:06:27Hillshum_ow long do i bridge it?
18:06:49dany_21aafter plugging in - wait some seconds and then take away the bridge
18:07:02dany_21ait mostly works... sometimes i have to repeat that procedure
18:07:11dany_21aafter that check dmesg
18:07:16Hillshum_or steal hub from sister
18:07:18dany_21aor "sudo fdisk -l"
18:09:37 Quit perrikwp (" ajax IRC Client")
18:10:28 Join faemir [0] (
18:12:30 Quit Hillshum_ ("CGI:IRC (Ping timeout)")
18:15:39 Quit MethoS-- (Remote closed the connection)
18:18:31 Quit dabujo (Read error: 104 (Connection reset by peer))
18:18:52 Join Hillshum_ [0] (
18:19:05 Nick fxb__ is now known as fxb (
18:23:56 Join DerPapst [0] (n=DerPapst@
18:24:57 Join Hillshum__ [0] (n=4ba5e7c0@gateway/web/cgi-irc/
18:25:39 Join MethoS- [0] (
18:26:12Hillshum__had to reboot
18:26:38 Join perrikwp [0] (i=98213910@gateway/web/ajax/
18:26:42 Join bertrik [0] (
18:27:22Hillshum__plugging it in normal dosn't show up on hardy live
18:30:43Hillshum__never mind it was in mtp
18:30:51 Quit petur ("work->badminton")
18:33:34 Quit Hillshum_ ("Leaving.")
18:34:58 Join herrwaldo [0] (
18:35:54 Join Strife89 [0] (n=michael@
18:36:14Strife89How's it looking for the release?
18:37:41amiconnAre there opinions on whether mp3-on-cop and the a52 speedup should be put into the 3.0 branch?
18:38:18gevaertsI wouldn't put mp3-on-cop in 3.0. No opinion on the a52 speedup
18:38:38 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:39:10markunmaybe we can release a 3.1 soon with all these new and untested improvements
18:39:25Strife89I like that idea.
18:39:56LambdaCalculus37Me too.
18:40:07LambdaCalculus37Strife89: Tonight's the night!
18:40:44Strife89What time? (just curious)
18:41:49LambdaCalculus37Strife89: According to Bagder, 23:00 his local time.
18:42:20Hillshum__+ 1 or 2 or so
18:42:30Strife89GMT -4 for me, I think.
18:43:33 Join kokoon [0] (
18:43:36kokoongood day
18:44:05LambdaCalculus37Strife89: No, you're GMT -5, like I am.
18:44:18kokoonare there any devs with CFmod knowledge here?
18:44:34Strife89LambdaCalculus37: My bad; I got confused.
18:44:40LambdaCalculus37kokoon: Ask away. Anyone who knows the answer can help you.
18:44:43kokooni could use some help...
18:44:47kokoonok sure
18:45:16kokoonso i've been running a Transcend 32Gb CF card in my Iriver H320 for a while now
18:45:32kokoonand i got tired of all the I03 errors a few times a day
18:45:47kokoonand i thought something must have changed in the last 3 months (or so)
18:46:06kokoongot the newest bootloader - or at least i think i did
18:46:06 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:46:17kokoonand now i'm stuck with ata -80 error
18:46:23kokoonend of story
18:46:32kokoonso from what i've gathered
18:46:35Hillshum__dany_21a: ivve run dmesg and the sansa connects but disconnects before it finnishes
18:46:37Strife89I had an idea earlier today, and it stems from the fact that I have found a lot of music (mostly game remixes, I'll admit) that is 100% free. So, for those looking to expand their collection, perhaps we could have a FreeMusic Wiki page?
18:46:42 Join childofkorn [0] (
18:46:48kokoonthe bootloader i was using before was patched
18:47:04kokoonand those patches fixed that exact issue
18:47:17kokoonam i right?
18:48:03kokoonoh, forgot to say, i also got a reply on the forum that said that my I03 problem was from a bug in a bootlader version from back then (that i used)
18:48:14kokoonso... there's still no question, is it...
18:48:18LambdaCalculus37Strife89: Perhaps, but why don't we also ask about it in -community?
18:48:21kokoonhow do i make it work?
18:48:39Strife89LambdaCalculus37: I'm in there, hang on and I'll paste.
18:49:18childofkornhi - i found a bug - and got a workaround for me, can anyone please tell me where to report it properly?
18:50:08domonokychildofkorn: if it really is a bug, put it into our flyspray Bug/patch tracker... but perhaps first tell us the problem :-)
18:50:16*amiconn will now test bootboxes and then start building flash images
18:50:27 Nick advcomp2019__ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
18:50:42dany_21aHillshum__: have you tried to repeat it a few times?
18:50:47amiconnI could need a tester with an fm recorder or a recorder v2
18:50:59amiconnThe others I can test myself
18:51:26markunkokoon: I don't know myself, but I've seen other people about CF problems today
18:51:46domonokyamiconn: have you seen fs#9411 ? (about flashing problems on recorderfm)
18:51:54kokoonyou mean you know who can help
18:52:03kokoonmarkun: or did you have problems yourself
18:53:49amiconndomonoky: No, but I just checked. Maybe he's running an old flash image. I have no idea without further information
18:53:51markunkokoon: maybe petur knows something about it, but he's not here right now. He's here pretty much every day.
18:54:05childofkorni update relatively often to new rockbox builds. i have an iriver ihp 120 on which i had installed the iriver firmware 1.66E. after the last two rockbox updates my mp3 playing was nearly a complete mashup since i got a distorted signal on my right earphone. downgrading to 1.63 fixed the problem
18:54:13 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
18:54:28markunamiconn: do you know anything about it?
18:54:36kokoonmarkun: yeah, i saw that nickname, you think he'd be back today?
18:54:38domonokychildofkorn: thats known, and already fixed in the latest builds, afaik..
18:54:55markunkokoon: could be, probably having dinner now.
18:55:01kokoonah cool
18:55:03kokoonthanks a lot
18:55:47domonokyamiconn: i talked with this user, and he had the new build roloed when he tried to flash, but perhaps he needs to update the whole flash (not only one image) ?
18:55:52kokoonbesides that... is there maybe something like a "mother of all" CFmod wiki pages/ forum posts?
18:55:59kokoonthe info is scattered everywhere
18:56:07 Quit Hillshum__ ("CGI:IRC (Ping timeout)")
18:56:17amiconndomonoky: That's what I mean. He doesn't state any relevant version information
18:56:18childofkorndomonoky: strange, because it was today when i fixed it myself... on the other hand - can't it be because of the bootloader? because i never update the bootloader itself but the build.
18:56:55amiconnBtw, bootbox has grown so that we need to cut the next 4KB flash block from main image space. :(
18:57:17domonokychildofkorn: the distorted sound on h1x0 was a problem with the rockbox build, not bootloader or iriver firmware..
18:58:33childofkorndomonoky: which was fixed not in the immideately following build? because i updated on a slightly newer one one day later and still had the problem (maybe 2 days ago)
18:59:21domonokychildofkorn: as far as i know it was broken for a few days, as nobody noticed the problem :-)
19:00:12childofkornwell, i am a hardcore - mp3 and software user
19:00:26childofkornok, fixed for me - never mind. i guess i will not upgrade the iriver firmware anyway, because it did not bring _any_ advantage for me
19:01:19 Join Hillshum__ [0] (n=4ba5e7c0@gateway/web/cgi-irc/
19:02:01 Quit Hillshum__ (Client Quit)
19:03:44amiconnWe certainly don't need arabjoin and hangul handling in the bootloaders, do we?
19:04:26LambdaCalculus37I don't believe so.
19:04:48childofkornone more for the kill? - battery is a topic to talk years about, but, when rockbox shuts down because of lack of battery power, i frequently start the iriver fimware, because there is a little more "power" left, since it recalculates the power a little faster on startup - here it comes: i get no more files and folders, and the only entry i see is completely shredded. can it be that rockbox can not unmount the filesystem proper
19:05:23childofkornplease excuse my english 0_o
19:06:13domonokychildofkorn: perhaps the OF doesnt show any files, because there is not enough power to spinup the HD ? (but we dont care about the OFs here, so thats offtopic) .. :-)
19:06:13 Quit Hillshum (Read error: 104 (Connection reset by peer))
19:08:55childofkorni can slightly deny; i pluged it in just to turn rockbox regularly on and off, that the shutdown is no longer battery caused and plugged it off imediately. after that i can access all files and folders with the OF
19:09:28childofkornif rockbox' shutdown causes the problem, it is close to the border of OT
19:13:17 Join Hillshum [0] (
19:13:47childofkornrechage time was just as long as both, startup and shutdown needed to properly happen. so this was my first guess. since i am really into IT- I wanted to clear out as much as possible
19:17:51childofkornany idea?
19:23:02***Saving seen data "./dancer.seen"
19:26:24amiconnDisabling arabjoin and bidi code saves ~2KB uncompressed, ~1.4KB compressed
19:26:29 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:26:54amiconnI'll also disable greyblitting for the bootloaders
19:27:43amiconnToo bad that the binsize table shows no bootloader stats
19:29:47Hillshumdany_21a: success! switched to .16F (
19:30:27 Join miepchen^schlaf [0] (
19:32:03gevaertsmarkun: can we assume that the first 32K of M6SL.ebn is the bootloader?
19:33:51 Join Llorean [0] (n=DarkkOne@
19:35:19 Join pixelma [0] (i=d5494d62@gateway/web/ajax/
19:35:21 Quit Hillshum ("Leaving.")
19:36:06 Join Hillshum [0] (
19:38:32pixelmaok, just committed some manual changes to the main trunk. If someone thinks it's worth putting them into the release branch as well (and could test compile for all release targets, though I've tested quite a bit myself) and has the time to commit them now, feel free...
19:40:00gevaertspixelma: I can test-compile them, so I'll have a go
19:40:21pixelmaI could only do this myself in a bit more than an hour or so, this might give some more time for a reaction before the release
19:40:30pixelmathanks :)
19:41:06gevaertsWe're talking about r18580 and r18581, right?
19:41:08amiconnugh, bootbox is buggy :(
19:41:41 Quit Hillshum ("Leaving.")
19:42:57amiconnI've just locked myself out from my recorder :(
19:43:08LambdaCalculus37I can't even get bootbox to work correctly on my Recorder.
19:43:22amiconnUSB doesn't work
19:43:44amiconn...and I have bootbox in both slots now, and no proper ajbrec.ajz on disk
19:44:55pixelmagevaerts: yes - if some people agree they're worth it
19:45:16gevaertsI do at least
19:46:01 Quit desowin (Excess Flood)
19:46:37amiconnpixelma: Do you think the cube.c changes should go into release? (fixes the impossible combos, but needs fixing the manual as well)
19:47:24pixelmagevaerts: nice, got to leave, catch my train
19:47:44pixelmaamiconn: don't know but would have prepared the manual change already ;)
19:48:08 Quit pixelma (" bbl")
19:51:54 Quit Darksair ("ERC Version 5.3 (IRC client for Emacs)")
19:52:32Strife89LambdaCalculus37: Finished the page, if you want to look at it.
19:53:26LambdaCalculus37Strife89: I made one little fix so that BitTorrent doesn't appear as a dead link.
19:53:37Strife89Whoops. :)
19:55:58gevaertsThe manual contains an empty changelog. Shouldn't we just drop that?
19:56:46markungevaerts: isn't the size of the booloader in this 'header' after the jump table?
19:56:47 Join Lear [0] (i=chatzill@rockbox/developer/lear)
19:57:34gevaertsmarkun: has my understanding of it
19:58:54childofkorndomonoky: i realized that i forgot to mention your name before i replied. did you read what i replied?
19:59:19markungevaerts: looks right
20:01:58markungevaerts: hm, wait, the bootloader doesn't contain the LCD init, does it?
20:02:17gevaertsmarkun: the "Loading" string is after the first 32k
20:09:50amiconnPhew, actually not locked out, just didn't press the correct button to get the alternate (working) bootbox
20:10:19 Join Siku [0] (
20:10:45amiconnI wonder what happened to bootbox though. It doesn't want to enter usb mode. Almost looks like a threading problem (some thread not acknowledging usb)
20:10:51 Quit perrikwp (" ajax IRC Client")
20:11:05amiconnSounds like a binchop task :\
20:12:13 Quit childofkorn ("Miranda IM! Smaller, Faster, Easier.")
20:12:24amiconnLambdaCalculus37: Btw, in what way(s) doesn't it work correctly for you?
20:15:44LambdaCalculus37amiconn: I flashed bootbox to my JBRv1, but nothing works. The charging screen doesn't appear when I hook it to the AC adapter, and the rescue mode (F1+ON) doesn't work, either.
20:16:01*LambdaCalculus37 probably thinks it's because he didn't flash something properly
20:17:16amiconnWell, if you flashed the .ucl, it will boot as the standard image. F1 is for the alternate one, i.e. the old (working) bootbox
20:17:30amiconnCharging screen works for me, as does loading ajbrec.ajz
20:17:57amiconnBut if ajbrec.ajz isn't there, it asks to use usb to fix it, but then fails to enter usb mode
20:18:02amiconnThe old one works
20:18:35amiconnIt also fails to enter usb mode when detecting early usb
20:18:46LambdaCalculus37But is that flashing rockbox.ucl or rombox.ucl?
20:19:03amiconnThere is no rombox for bootbox
20:19:38 Quit domonoky (Read error: 104 (Connection reset by peer))
20:20:27LambdaCalculus37amiconn: I flashed bootbox to my Archos JB5000. Then I "played" rombox.ucl afterwards and flashed that.
20:20:58LambdaCalculus37But what about on the Recorder?
20:21:11*amiconn doesn't understand what LambdaCalculus37 is trying to do
20:21:22amiconnI thought you wanted to test a current bootbox
20:22:07 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:22:20 Join Schmogel [0] (
20:23:13 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
20:23:48*LambdaCalculus37 starts from the top again
20:24:36LambdaCalculus37Okay, I just tried F1+ON on my JBRv1 again. It's working.
20:24:50LambdaCalculus37I think I was just doing the button combination wrong.
20:25:11amiconnF1+On starts the alternate flash image, which is the old, stable bootbox
20:25:25amiconn..which in turn will rolo an ajbrec.ajz if present
20:25:40LambdaCalculus37amiconn: I think I got confused by the wiki and manual instructions, which are horribly outdated at this point.
20:25:50 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
20:25:50amiconnIf you want to test a current bootbox without risk, you need to flash it like you would normally flash rockbox
20:26:09amiconnI.e. "play" rockbox.ucl from a bootbox package
20:26:32 Quit homielowe (Read error: 110 (Connection timed out))
20:26:32amiconnWhen powering off & on afterwards, _this_ bootbox will start by default
20:26:59amiconnThen you can check what works and what doesn't, e.g. by renaming ajbrec.ajz to something else and rebooting
20:27:15amiconnBootbox won't find ajbrec.ajz and ask you to fix it via usb
20:27:42amiconnIf it doesn't work, you can still fix it via the alternate (old) boobox, by booting with F1 pressed
20:28:37bluebrotherboobox? ;-)
20:28:43amiconnOnce we can be sure that a current bootbox works, I will make new flash images from them
20:29:23amiconnI mean those things which are flashed using firmware_flash, and which will contain the new (fixed) bootbox as the alternate one
20:29:23 Quit Dementio (Read error: 104 (Connection reset by peer))
20:29:41 Join Dementio [0] (n=dementio@
20:32:20gevaertsmarkun: meizu force off is long play, right?
20:35:44 Quit moos ("Rockbox rules the DAP world")
20:40:11 Join matsl [0] (
20:44:06 Quit parafin ("So long and thanks for all the fish")
20:44:21 Join parafin [0] (
20:53:00 Join {phoenix} [0] (
20:55:30 Join tucoz [0] (n=martin@rockbox/staff/tucoz)
20:56:24 Nick plus is now known as plus_M (
20:57:09*tucoz wonders what to do about the working with playlists chapter and the 3.0 release
20:57:19tucozfor the manual that is
20:58:24tucozThere is a fixme there right now. I would like to suggest that we get rid of that, and either keep the content as-is, or improve what is written
20:58:40 Quit pixelma2 ("-")
20:58:54 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
20:59:05tucozmost important not to have the fixme there for the release manual
20:59:54gevaertsI'd work on it, but I've never actually used playlists (except for "insert last"). How wrong is that chapter?
21:00:25tucozthere is also a fixme for trigtype in the recording settings chapter, which i think we should fix. not sure what that is thought.
21:01:17tucozgevaerts: i am not sure. i only used playlistst like yourself
21:01:27pixelmathere's a more informative page in the wiki about triggered recording, I wanted to move the info over one day...
21:01:38tucozbut i'll read through it and see if something obvious pops up
21:02:12 Join Seed [0] (
21:02:34tucozI think we can safely remove "more description is needed" from the doom chapter.
21:03:11gevaertsAt least from the 3.0 manual, yes
21:03:34tucozand also appendix D menu overview
21:03:40 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:03:43pixelmaI agree
21:03:55tucozok, i'll fix those two things right now
21:04:13gevaertsI'll remove the m:robe 100 warning about rbutil 1.0.5
21:04:24pixelmawhat happened to the "disabling broken plugins" plan?
21:04:47 Quit bertrik ("Leaving")
21:04:48gevaertsI disabled several of them
21:05:00pixelmain the manual too?
21:05:09amiconnIs there an easy way to find an (approximate) SVN revision for a specified date?
21:05:16*gevaerts checks to make sure
21:05:41*amiconn needs to find out how far to go back for binchopping
21:06:06bluebrotheramiconn: you can svn up -r <date>
21:06:18gevaertsamiconn: svn log -r '{2008-09-23}'
21:06:37pixelmagevaerts: for which players did you disable metronome? Need to update here first, and actually first really arrive here and grab something to eat
21:06:53*bluebrother just realized the log version himself :)
21:06:58bluebrotherjust a bit too late
21:07:08gevaertspixelma: c200, ondio and player
21:07:11amiconnGah, >8000 revisions to chop through...
21:07:32pixelmagevaerts: ah ok, those are the ones I thought about too
21:08:00gevaertspixelma: yes, you told me that list :)
21:08:11tucozamiconn: no commit log messages that can help you?
21:08:14 Quit Siku ()
21:08:22gevaertsouch, still in the manual
21:08:29 Quit miepchen^schlaf ()
21:08:32 Quit scoles ("CGI:IRC (Ping timeout)")
21:09:05amiconntucoz: How would that help? I have no idea in which source file the bug hides, and I know that bootbox from July 2006 works, and current bootbox is broken
21:09:24amiconn2006-07-1 is r10160
21:10:46gevaertsI'll fix the volume ranges in the manual
21:11:11tucozamiconn: i was thinking of searching through for clues, as you get a nice formatting and each revision's files next to each other.
21:11:39 Join miepchen^schlaf [0] (
21:14:39amiconnNot if I have absolutely no idea what could be the cause
21:15:07amiconnIt's *probably* something related to USB_ACKNOWLEDGE, but that's all I have atm
21:15:34tucozsounds hard to find out, but very cool if you do
21:15:49amiconnBinchopping has the advantage that it doesn't take that long even for large ranges
21:16:06amiconn8000 revisions should be 13 steps
21:16:39amiconnBut I'm running out of time; Bagder wants to start the 3.0 builds in about 2 hours
21:17:04gevaertsSome ipods have the default battery capacity as unknown in the manual. Should I fix those with whatever is in config-*.h, or do we need confirmation that it's also the real battery capacity?
21:17:09*tucoz found 22 results for binchop in google
21:17:26gevaertstucoz: also called bisect
21:17:27tucozsorry 65
21:17:30tucozi see
21:17:40Lloreangevaerts: Even if it's not the actual capacity, it's still the default, right?
21:17:48tucozthat i know of
21:18:01 Join mcuelenaere_ [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:18:30Lloreangevaerts: So I don't see any reason not to document it, maybe with a note saying (this value may not reflect the actual battery capacity in some cases) or something
21:19:02tucozwhat do you say about the changelog? Use MajorChanges from the wiki?
21:19:28*gevaerts looks
21:19:33Lloreantucoz: It's a great start, at least
21:19:49 Join AndyIL [0] (i=AndyI@
21:20:20tucozLlorean: ok
21:21:00pixelmagevaerts: thanks for committing my changes to the release branch
21:21:26LambdaCalculus37tucoz: I was going to add some more description to the doom manual page, but I haven't got commit access just right now, but I'll do it later on. Nothing fancy; just some basic gameplay description and the like.
21:21:34 Join PaulJam [0] (
21:22:21ameyerLlorean: also, with all supported iPods being 2+ years old and what lithium ion batteries do as they age, I'd say it certainly doesn't reflect the actual battery capacity, YMMV, though.
21:22:33pixelmaLambdaCalculus37: so that people get the right impression of Rockbox? It's all about doom... ;)
21:22:52tucozLambdaCalculus37: do you want it in for the 3.0 releas?
21:23:03***Saving seen data "./dancer.seen"
21:23:51Lloreanameyer: Well the goal is for the default to be the factory installed capacity. In a few cases this probably isn't true still.
21:24:38ameyerfair enough
21:24:50*ameyer should have kept his mouth shut
21:26:19LambdaCalculus37tucoz: No, not needed. And actually, pixelma has a point.
21:26:31LloreanBasically, the "default" is also the "minimum" in many cases, since it's rare people buy replacement batteries with lower capacities than the original.
21:26:38 Quit AndyI (Read error: 60 (Operation timed out))
21:27:14ameyerafter the release, there'll still be official svn builds, right?
21:27:38 Quit AndyIL ()
21:27:41LambdaCalculus37Why would we stop official SVN builds? ;)
21:27:49tucozshould the changelog be in ascending or descending order by date? or some other arrangement?
21:27:57gevaertsLlorean: the default settings aren't changed yet. I most probably won't do that as (a) I'm busy with the manual now, and (b) I've never touched settings code before, and I don't want to break the release
21:28:11ameyerLambdaCalculus37: because you want everybody to use the release?
21:28:18gevaertstucoz: I commented out the changelog in main.tex, so you'll have to re-enable it
21:28:25 Join AndyI [0] (i=AndyI@
21:29:09tucozgevaerts: yes i noticed
21:29:31ameyeralthough having development builds avaliable for testing has its benefits
21:30:01amiconnameyer: If you ask me, the release is just a point in svn history. Especially the current one...
21:30:21*amiconn thinks the release doesn't have release quality, even
21:30:48*ameyer agrees and points at the sansa/iPod usb reboot bug
21:31:01 Join scotty_007 [0] (
21:31:07*gevaerts somewhat agrees, but thinks that it's good to get releases going anyway
21:31:24*tucoz agrees with gevaerts
21:31:31amiconnThere are dozens of gui glitches (wrong fonts, scrolling lines from old screen poking through, wrong list behaviour...)
21:31:34*LambdaCalculus37 also agrees with gevaerts
21:31:59*gevaerts wonders what happened to r18459
21:32:12bluebrotherwell, it's a .0 release. Doesn't give any excuse of being bad, but anyway ...
21:32:39tucozand it is also the first release in 3 years
21:32:57bluebrotherthe problem I see is that after the release everything goes back to "normal". It would be good to have more focus on bugfixing in the next months though / instead.
21:33:12bluebrotherto get a better 3.1
21:33:48amiconnhuh??? Even r10160 bootbox doesn't werk when built now??
21:34:03Lloreanamiconn: Has the GCC version we use changed?
21:34:07gevaertsLambdaCalculus37: r18471 overwrote some of my earlier changes
21:34:09amiconnyes it did
21:34:32pixelmait's interesting to see how the set release date and time seems to motivate people... now
21:36:15 Quit mcuelenaere (Read error: 110 (Connection timed out))
21:36:42LambdaCalculus37gevaerts: That would be my fault. :P What did I break?
21:36:47Leargevaerts: r18459 was on the branch...
21:36:53gevaertsLear: yes
21:37:06gevaertsLambdaCalculus37: disabling metronome and md5sum
21:37:13tucozbluebrother: i have forgotten my latex skills. what environment do you think i should use for the changelog?
21:38:02bluebrothertucoz: do we want icons similar to MajorChanges?
21:38:25LambdaCalculus37gevaerts: Oops. Sorry about that.
21:38:45pixelmaLambdaCalculus37: ah, there it was when dice was put into the "games" section... I'm not sure if it really belongs there - maybe the manual was more right before
21:39:32LambdaCalculus37pixelma: I put dice into the "games" section since the plugin is located in .rockbox/rocks/games.
21:40:15 Quit J-23 (Read error: 104 (Connection reset by peer))
21:40:24pixelmabut stating as it is now is probably the quicker solution. I did that with the button tables when I started thinking "this button map isn't nice"
21:41:06pixelmaLambdaCalculus37: I know, but you can discuss if dice is really a game (IMO it's not currently)
21:41:31pixelmabetter at another time though
21:42:03 Join tucoz_ [0] (n=martin@
21:42:37 Quit tucoz (Nick collision from services.)
21:42:46 Nick tucoz_ is now known as tucoz (n=martin@
21:42:53 Quit mcuelenaere_ (Read error: 110 (Connection timed out))
21:43:02LambdaCalculus37pixelma: If it isn't a game, we should reclassify it then.
21:43:10tucozbluebrother: sorry, the current went off
21:43:22bluebrothernoticed you vanishing ;-)
21:44:05bluebrotherwell, you could use the description environment and put icons in the description, like \star
21:44:06tucozbluebrother: i was thinking of arranging the different changes in three sections. Enhancements, New features, bug fixes
21:44:17bluebrotherhaven't tried yet, but that could work
21:44:25bluebrotherin that case, why not simply use enumerate?
21:44:35pixelmaLambdaCalculus37: yes, that's what I ment. Debatable though
21:44:36tucozright. i'll try that. :)
21:44:38bluebrotheror better, itemize? No need for numbers ;-)
21:45:19tucozbut i do no think that the different items should be emphasized as bold.
21:45:27bluebrotheranother way would be to use itemize and make \star the bullet ;-)
21:45:48LambdaCalculus37pixelma: The way I see it, is that since dice are used to play games, then the dice plugin should be classified as a game.
21:46:06LambdaCalculus37But I know that since dice isn't a game per se, some would say that it should be classified as an app.
21:46:32*Llorean thinks it's an app.
21:46:34pixelmamake a new category - gameapp ;)
21:46:56*gevaerts thinks it's useless, and therefore a demo
21:47:17pixelmabut actually I didn't want to start that discussion now...
21:48:27*bluebrother goes trying something
21:49:26 Quit scorche (Read error: 104 (Connection reset by peer))
21:50:17 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
21:53:21 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
21:55:11 Join vitja [0] (n=vitja@
21:57:10gevaertsMain fixmes remaining : archos-flashing, working_with_playlists, Trigtype. There are also some minor ones in config_file_options where some ranges or step sizes are missing
21:58:12 Quit scorche (Nick collision from services.)
21:58:20*amiconn seems to have found sth...
21:58:32 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:58:47gevaertsMy proposal for config_file_options is to just drop the fixme and just leave the config option less detailed
21:59:01 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
21:59:09LambdaCalculus37If the instructions listed in archos-flashing are valid, then I'll just drop the fixme and correct any spelling or lanuage errors.
21:59:16amiconnThe breakage of bootbox seems to have happened significantly earlier than when the current flash rom images were published
21:59:32LambdaCalculus37In the meantime, go home time is now.
21:59:34*LambdaCalculus37 leaves
21:59:35 Quit LambdaCalculus37 (" ajax IRC Client")
22:00:05gevaertsamiconn: so the current images work by chance?
22:00:15tucozregarding the changelog, should we list changes such as: "Removed the non-working hardware equalizer for iPod Video. Provide working bass/treble gain and cutoff settings instead."?
22:00:36gevaertstucoz: ipod was not released in 2.5, so...
22:00:44amiconnNo, they seem to contain a bootbox which was compiled significantly earlier than the images were released
22:00:45tucozgevaerts: yes, exactly.
22:01:21gevaertstucoz: I'd ignore those. Just list generic news, changes to archoses, and new players
22:01:23tucozso i do not add that to the changelog then
22:01:26amiconn*If* I am right, it was LinusN who broke bootbox, back in r8168...
22:02:10 Quit Nibbler ("Ex-Chat")
22:02:20tucozgevaerts: so improved power management for portalplayer should also be left out then...
22:02:38gevaertstucoz: I'd say yes, but maybe others disagree
22:03:17tucozlot's of changes will be left out then. like metadata-on-buffer, as swcodec didn't exist back then
22:04:20*gevaerts waits for other people to comment
22:04:22tucozbut i think in a changelog 2.5->3.0, details like that should be left out.
22:04:47 Join MethoS-- [0] (
22:05:12 Quit Dementio (Read error: 110 (Connection timed out))
22:05:48 Join miepchen^schlaf_ [0] (
22:07:58*amiconn was right (!)
22:08:50 Quit mcuelenaere (Read error: 60 (Operation timed out))
22:08:50amiconnH300 bootloader usb broke bootbox
22:09:03 Quit {phoenix} (Remote closed the connection)
22:09:03amiconnBootbox needs the usb thread running
22:09:34pixelmaisn't that the one that causes trouble anyway?
22:09:44pixelmaon some H300s I mean
22:10:04gevaertsserves them right. They shouldn'd have broken bootbox
22:10:22 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
22:10:28amiconnLinusN's commit disabled the usb thread if BOOTLOADER is defined
22:10:59amiconnBut then re-enabling it causes bootbox to grow by 700 bytes :\
22:11:59pixelmablah, referenced the wrong revision (off by one). Sorry.
22:12:18*amiconn wonders why LinusN implemented H300 bootloader this way instead of using the thread
22:12:26*amiconn summons LinusN
22:13:07amiconnPerhaps no tick in the H300 bootloader?
22:15:00 Quit miepchen^schlaf (Connection timed out)
22:15:34*bluebrother got it :)
22:19:52pixelmagevaerts: I take it there were no problems with my earlier changes to the manual (just want to be sure that my testing is right)?
22:20:15gevaertspixelma: right. All 23 manuals built fine
22:20:40 Quit MethoS- (Read error: 110 (Connection timed out))
22:20:48pixelmathanks for the info :)
22:22:15 Quit bluebrother (Nick collision from services.)
22:22:20 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
22:22:28bluebrothertucoz: what do you think about a list formatted that way?
22:22:34bluebrotherit only requires a new package, but that should be part of a usual latex installation (it's part of tetex at least)
22:22:45bluebrother(looks like I got disconnected this time :o )
22:22:54gevaertsI just read the Working with playlists chapter, and I don't see anything wrong with it (except one typo). Shall I just remove the fixme?
22:23:06*gevaerts tried everything he wasn't sure of
22:23:22tucozbluebrother:looks great. i actually do it like that now
22:24:07bluebrotherI made up a new environment for that actually ...
22:24:49bluebrothercheck this:
22:25:30pixelmagevaerts: maybe someone forgot to delete the fixme (though I didn't look myself yet)
22:25:46tucozbluebrother: nice. do i add items with \item?
22:25:48gevaertspixelma: it also says it's probably incomplete. That may still be true of course
22:26:10bluebrothertucoz: yes, you use it exactly like itemize −− just \begin{changelog}. Should I commit it?
22:26:22bluebrotheror do you want to do it "one in all"?
22:26:28gevaertsIf it's only that, I think that's not serious enough to leave the fixme for 3.0
22:26:32tucozbluebrother: please commit
22:27:05tucozto the 3.0 branch as well
22:28:00bluebrothertucoz: done on trunk, branch in a minute ...
22:31:58gevaertsNow we only need a recording specialist to explain what Trigtype is
22:32:03bluebrotherand branch done as well.
22:34:46 Join kronflux [0] (
22:35:39 Join MethoS_ [0] (
22:37:43bluebrotheryou're welcome
22:39:23bluebrotherwow, the commit traffic on the manual is ... amazing tonight :)
22:41:13 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:43:49tucozhmm. should i just commit what i have done so far? I do not think i'll be able to finish this tonight.
22:43:58Bagder17 minutes...
22:44:25gevaertsBagder: do you know anything about recording? ;)
22:44:50tucozI have copied what i found suitable from the MajorChanges page, but there are some major things missing. Like some of the codecs, new targets etc.
22:45:41gevaertstucoz: I'd just list supported codecs. Targets are irrelevant, as the manual is target-specific anyway
22:45:57tucozwell, the changelog is not at the moment
22:46:02 Join BigBambi_ [0] (n=Alex@rockbox/staff/BigBambi)
22:46:07tucoztarget specific that is
22:46:18*gevaerts fears that unless a recording specialist turns up soon Trigtype will remain undocumented
22:46:44gevaertstucoz: you can always list all supported targets as well of course
22:47:06tucozi'll commit what i have for now
22:47:42*amiconn needs to squeeze bootbox by ~1.7KB somehow
22:48:10amiconnCompressed size, that is
22:48:27 Join mcuelenaere_ [0] (n=mcuelena@rockbox/developer/mcuelenaere)
22:48:50 Join LambdaCalculus37 [0] (
22:49:02tucozgevaerts: could you take a peek at the changelog?
22:49:10 Join amdgoon_ [0] (n=amdgoon@
22:49:12tucozi committed it to the 3.0-branch
22:49:19amiconnIf not, new flash rom images would have the start of the main image shifted by 4KB, meaning 4KB less room for it, and also that an older rombox wouldn't be compatible with the new rom image and vice versa
22:49:44 Quit MethoS-- (Read error: 110 (Connection timed out))
22:50:32gevaertstucoz: I think we need to cut everything that's not in 3.0, like unfinished ports and md5sum
22:50:36amiconnDoing USB "by hand" like the iriver bootloader would help a bit, but only that (a bit...)
22:51:10tucozgevaerts: yes. problem is i do not know what is supported...
22:51:17gevaertstucoz: I'll have a go
22:51:58pixelmagevaerts: your expert is here . I actually wanted to move that over to the tex (including some graphics that show the trigger bar, probably only black and white though). Unfortunately I think that'll take some time
22:52:10tucozgevaerts: what do you think about removing the lines where codecs are mentioned, and add those to a subsection of their own? \subsection{New codecs}
22:52:53tucoznot mentioning dates, as that will be time consuming to track down
22:53:24*LambdaCalculus37 makes a small fix to tools/configure; some comments were left on the c100 configure section
22:54:16gevaertstucoz: I committed my removals of not-in-3.0 things, so you can do the new codecs :)
22:54:36tucozcool thanks
22:54:51gevaertsLambdaCalculus37: can you check the Flashing chapter, or should we just remove it?
22:54:59gevaertsYou have 6 minutes
22:55:21LambdaCalculus37gevaerts: Checking now.
22:56:02pixelmaI guess that's helplessly outdated, nothing you can do in 6 minutes
22:56:07LambdaCalculus37gevaerts: We should just take the fixme out.
22:56:42amiconnImho we could update the release manuals even after the release
22:56:45*LambdaCalculus37 curses Aquamacs Emacs for being so bloody slow
22:56:57*gevaerts agrees with amiconn
22:57:04*Llorean agrees with gevaerts and amiconn
22:57:09*pixelma was thinking the same
22:57:14*LambdaCalculus37 agrees with Llorean, gevaerts, amiconn and pixelma
22:57:19gevaertsLet's just remove the Flashing chapter for now, until someone has had time to go over it
22:57:32LambdaCalculus37gevaerts: I'm still fixing up tools/configure.
22:57:37*gevaerts will do it
22:57:46pixelmago for it
22:57:55Bagderanyone besides LambdaCalculus37 gonna change any code the next 3 minutes?
22:57:58 Quit mcuelenaere (Read error: 110 (Connection timed out))
22:58:24tucozBagder: me
22:58:30tucozi'll be done in 5 minutes
22:58:36tucozif that is ok
22:58:38pixelmagevaerts too, I guess
22:58:42tucoznot code, manual
22:58:43Bagderthat's cool
22:58:48LambdaCalculus37Bagder: Committing now!
22:58:49gevaertsJust one manual commit left
22:59:14Bagderok, the last code commit then so I can now officially start my binary builds
22:59:53Bagderwell, after LambdaCalculus37's commit
23:00:20Bagderyou gonna do that in the 3.0 branch too?
23:00:39BagderI guess its not necessary?
23:01:15tucozBagder: wait
23:02:46Bagderbuilds now in progress
23:02:50*Llorean cheers
23:03:13*LambdaCalculus37 brings out several crates of beer!
23:03:16*tucoz dances the v3.0 dance
23:03:22*gevaerts goes to find beer as well
23:03:34*BigBambi_ already has beer
23:03:41BigBambi_Congrats to everybody
23:03:48gevaertsBigBambi_: don't drink and edit manuals!
23:03:59*LambdaCalculus37 pours several large steins of beer and passes them all around
23:04:19scorchedid we ever settle on a code name? ;)
23:04:20BigBambi_gevaerts: I know, it is dangerous. Thankfully I was sensible and made a choice - and beer won
23:05:19LambdaCalculus37scorche: I thought we were doing beer names. :)
23:05:43*gevaerts wants "coal powered release"
23:06:15BigBambi_Powered by the sweat of the innocent?
23:06:24*LambdaCalculus37 wants "Duvel"
23:06:24Bagder"another leap year, another release"
23:06:57 Quit Seed (Read error: 104 (Connection reset by peer))
23:07:11*pixelma read on the way home that today is supposedly the Linus' name day
23:07:17Llorean3.0: Still not supporting iPod Classic
23:07:32 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
23:07:41 Join Seed [0] (
23:07:53gevaertsLlorean: why not just Rockbox Classic?
23:08:00Bagderwrong name in the build script, restarting...
23:08:05Lloreangevaerts: No, that's 4.0
23:08:18Llorean3.0 can be "Rockbox Video"
23:09:10 Quit BigBambi (Read error: 113 (No route to host))
23:09:11*Bagder pats ccache on his head
23:11:21 Quit miepchen^schlaf_ ()
23:11:50bluebrotherwe already had "mayday" like a couple of years ago :)
23:12:25Bagder2.5 was released on sep 22
23:12:26*preglow feels big things coming
23:12:41gevaertsWe're a day late!
23:12:41preglowthree years, half a version
23:12:52ameyer"that only took 3.0 years"?
23:12:58scotty_007three cheers for all the awesome rockboxers!
23:14:04tucozis it 26 targets now?
23:14:49 Join homielowe [0] (
23:15:04 Quit domonoky (Read error: 104 (Connection reset by peer))
23:15:18*LambdaCalculus37 decides to go have a beer for real and heads to his favorite pub
23:15:20Bagderit depends on how you count them
23:15:40LambdaCalculus37See everyone later!
23:15:48 Join miepchen^schlaf [0] (
23:15:56 Quit LambdaCalculus37 ("Do quit now, there's a demon around the corner!")
23:16:05ryanakcaDidn't see it on the FAQ, so here it goes... My iPod's screen cracked, so I'm probably going to be replacing it by another music player in the near future... what make/model best supports rockbox?
23:16:27tucozok. everything on the current build page except the m3 and ipod 1g/2g?
23:17:08*ameyer guesses the H1xx and H3xx ports probably are the most mature swcodec ports
23:17:20gevaertsryanakca: it all depends on what you need
23:17:37*bluebrother suggests the BuyersGuide wiki page
23:18:01ryanakcabluebrother: thanks
23:18:23Nico_Phas the release notes wiki page been updates?
23:18:26amiconnHmm, disabling loadable font support (incl. font caching) for the bootloaders should be a good idea, shouldn't it?
23:18:37Bagderamiconn: yes!
23:18:38gevaertsamiconn: indeed.
23:19:11amiconnAt least it helps dropping bootbox size by ~1KB compressed, ~1.2KB uncomprseed
23:19:36 Quit Lear ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
23:20:00amiconnBagder: Any chance that we'll see binsizes for bootloaders as well?
23:20:45Bagderis the rockbox-info thing working for them?
23:21:29amiconnBinsize is, ram usage is not (shows 0)
23:21:50amiconnThat's for bootbox, I don't know whether it applies to the others
23:21:55Bagderit will then "just" require that the master gets the bootloader zips too
23:22:18 Quit mcuelenaere_ (Connection timed out)
23:22:51 Quit ompaul (Client Quit)
23:22:59amiconnRAM usage probably doesn't tell much, as the bootloader ram layout is quite different between targets
23:23:06***Saving seen data "./dancer.seen"
23:23:57 Part tucoz ("Ex-Chat")
23:26:12DerPapstCongrats to all for the 3.0 release :D
23:26:55Nico_Phas it been made yet?
23:27:35DerPapsti thought bagder is compiling it already :P
23:27:39BagderI am
23:27:47Bagder29 zip files done
23:27:53DerPapstw00t ;)
23:28:16 Quit Seed (Nick collision from services.)
23:28:19 Join Seedy [0] (
23:29:03*Bagder copies zips to download mirror master
23:29:27 Join homielowe_ [0] (
23:29:59 Join homielowe__ [0] (
23:30:07 Part amdgoon_
23:30:17 Quit homielowe__ (Client Quit)
23:30:56Bagdertime to fix up the download page
23:31:21*DerPapst is already courious about the next months statistics :)
23:31:21*Llorean cheers
23:31:43preglowwish i had a beer :/
23:31:48 Join Neovanglist [0] (i=Neovangl@
23:32:20 Join einhirn [0] (
23:32:43scotty_007damn, ipod is busy with battery bench (mp3-cop-test) - no chance to install 3.0 for the next hours :-(
23:32:47 Join lasser [0] (
23:32:47*DerPapst dccs preglow a can of beer
23:33:45dionoeaBagder: tell me if you need to force a mirror sync
23:34:15Bagdernah, the other two won't get that anyway so we can just let time fix it
23:35:14dionoeabtw, not sure if you noticed last week but the videolan server had some downtime due to our new VLC release. We're likely to release a bugfix version tomorrow so I hope we won't have any more downtime.
23:35:25 Quit homielowe_ (Read error: 60 (Operation timed out))
23:35:31dionoeaand congrats on the release ;)
23:36:52 Join LambdaCalculus37 [0] (
23:38:03preglowanything in the way of release notes?
23:38:07kronfluxwelcome LambdaCalculus37!
23:38:26 Quit homielowe (Read error: 104 (Connection reset by peer))
23:38:46dionoeaBagder: updating the home page before the download page seems kind of backwards ;)
23:38:50LambdaCalculus37kronflux: Just in time! Grab a beer and celebrate with us! Rockbox 3.0 is on the way!
23:38:56pixelmadionoea: ah, guess that's why some menauls weren't available for me some days ago
23:39:08dionoeapixelma: probably ;)
23:39:08*pixelma glares at typo...
23:39:16Bagderdionoea: yeah yeah but in a few minutes all that won't matter ;-)
23:39:43Bagderthe download page is there now
23:39:57Bagderjust pointing to zip files not present yet ;-P
23:40:03gevaertsBagder: weren't we going to drop m3 from the list?
23:40:05amiconnBagder: As soon as I manage to get both the size to behave, and usb to work (these are related problems), I would like to make new archos rom images
23:40:23Bagdergevaerts: we are, just my mistake
23:40:35amiconnBut then, even if I build them from the 3.0 branch, they're technically no longer equivalent to 3.0 code, because I need to fix things first...
23:40:56bluebrothermake a 3.0.1 release for them?
23:41:07 Join tvelocity [0] (n=tony@
23:41:41gevaertsbluebrother: not really needed, as they don't replace any 3.0 files
23:42:24Nico_Pnow we need a few announcements!
23:42:28gevaertsI'd keep 3.0.1 for when a build that got released as 3.0 gets replaced
23:42:43bluebrotherhmm, rbutil needs some small adjustments for 3.0 downloads
23:42:45amiconnBagder: Oh, and do you know by chance when LinusN will be around again?
23:42:48preglowdoesn't matter if it's technically not the same code base, as long as it looks like it is
23:43:03bluebrotherwe need to make that retrieve this information from the server too ...
23:43:13Bagderamiconn: no I don't, I think he's just being busy with life stuff
23:43:18 Join dan_a [0] (n=dan-mirc@
23:43:21Nico_PBagder: did you tag the svn rev you compiled?
23:43:39dan_aCongratulations on the 3.0 release everyone!
23:43:44*gevaerts looks at the download table and thinks that dropping the m3 was aesthetically the right decision
23:43:53 Quit matsl (Read error: 110 (Connection timed out))
23:43:54kronflux3.0? yay!
23:44:00Bagder18607 it is though
23:44:00bluebrotherIpod 1G2G was a release target?
23:44:00gevaertsNico_P: we can still do that I think
23:44:03Lloreanpreglow: On the other hand, we might want to grab the 3.0.1 number before someone tries to make an unsupported build named that. :-P
23:44:13gevaertsbluebrother: it wasn't explicitely dropped
23:44:13Nico_Pgevaerts: yes. it should be done
23:44:26bluebrotheryou can tag any arbitrary revision
23:44:41preglowLlorean: grab "rockbox ultimate" and "rockbox extreme" while you're at it
23:44:50preglowalso "rockbox home" and "rockbox enterprise"
23:45:04*DerPapst grabs berrockbox
23:45:39DerPapstnot that i'm going to maintai an unsupported build :P (too lazy)
23:45:53Bagdersell the name!
23:45:57kronfluxLambdaCalculus37: mind you, I cant celebrate fully. I've no mp3 player which can be hacked yet.
23:46:46*bluebrother schedules some small rbutil adjustments for the next days
23:47:04 Quit Seedy (Read error: 110 (Connection timed out))
23:47:26*DerPapst donates the name to Bagder :)
23:48:15gevaertsNico_P: rockbox-info.txt has the exact revision, so it's not that hard to find
23:48:17LloreanBagder: Do we want to try to set a date for the 3.1 freeze or better yet, set a date for setting a date, such as saying "We want to do 3.1 in Month" then when we get to month see what period seems best"?
23:48:21*LambdaCalculus37 sees the new relase page!
23:49:20Nico_Pgevaerts: still, I think it's a good practice to tag releases. that way you can do an svn co svn://
23:49:25 Join matsl [0] (
23:49:59BagderLlorean: I'd like a 3 month period to the next freeze
23:50:19LloreanI was thinking late Dec early Jan-ish
23:50:27gevaertsNico_P: I don't disagree, but I think that whoever did the branching and release building should do that
23:51:20Bagderthere's a 3.0 in tags/ now
23:51:25ameyerI don't count, but maybe the middle of December?
23:51:25Bagderv3_0 in fact
23:51:35 Quit reacocard (".")
23:51:54bluebrotherwhy not v3.0? It's just a folder name (unlike cvs)
23:52:12BagderI just followed the existing naming convension
23:52:12 Quit Llorean (Read error: 104 (Connection reset by peer))
23:52:57 Join Strife1989 [0] (n=michael@
23:52:58bluebrothergood point.
23:53:17amiconnHmm, bootloaders don't need ata idle callbacks either, do they?
23:54:14Bagderso is there any further 3.0 manual work planned or should I make a set of those too?
23:55:40gevaertsBagder: I'm not planning anything right now
23:55:50 Quit jgarvey ("Leaving")
23:56:22*Bagder invokes script
23:56:40dionoeahum, isn't the offical install method using rbutil nowadays?
23:56:54dionoea(I haven't followed all the prerelease discussions so sorry if i missed soemthing)
23:57:45gevaertsdionoea: rbutil and manual are equally "official", but we (or at least I) prefer users to use rbutil as it's less error-prone for people who aren't familiar with it
23:57:45 Quit jhulst (No route to host)
23:57:56Bagder has the best screenshot ;-)
23:58:08dionoeagevaerts: my question was about the download page in fact ;)
23:58:14dionoeait doesn't mention rbutil

Previous day | Next day