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 2008-07-06

00:00:35 Quit fyrestorm (Read error: 104 (Connection reset by peer))
00:00:57 Join fyrestorm [0] (
00:08:53 Quit sarixe ("(EE) Failed to load "quit" module")
00:09:16 Join sarixe [0] (
00:09:20 Quit petur ("Zzzzzz")
00:12:22 Join Thundercloud [0] (
00:12:25 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
00:18:26 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
00:22:59 Join desowin [0] (n=desowin@atheme/member/desowin)
00:23:46 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
00:23:46 Quit mcuelenaere (Remote closed the connection)
00:32:43 Quit PaulJam (".")
00:39:49 Quit sarixe ("(EE) Failed to load "quit" module")
00:43:22***Saving seen data "./dancer.seen"
00:43:53 Quit desowin ("KVIrc 4.0.0 Insomnia")
00:53:19 Join tvelocity [0] (
00:59:37 Join faemir [0] (
01:00:14 Quit safetydan ("Leaving")
01:01:06 Quit delorean90 (Read error: 110 (Connection timed out))
01:16:39*bluebrother has to fix several conflicts in his newly updated german translation file ;-'(
01:17:55saratogai just decoded a wma file in the sim using the Tremor IMDCT
01:19:37LloreanAnd how'd it go?
01:20:30bluebrotherlooks like running lupdate on non-touched languages isn't always a good idea.
01:21:04 Quit mirak (Read error: 104 (Connection reset by peer))
01:21:31saratogaLlorean: suprisingly it sounded normal
01:22:00LloreanYou mentioned the math seemed a bit weird for that one earlier, right?
01:22:55saratogayeah I had to dig up the paper explaining it
01:23:07saratogastill makes no sense to me, but the literature really seems like that algorithm
01:23:26saratogaand the paper starts by talking about how its better then the one used in wma/aac
01:24:43saratogatheres some minor distortion when used on arm, so i guess it won't be quite that easy
01:25:11 Join delorean90 [0] (
01:25:46 Quit Thundercloud (Remote closed the connection)
01:25:48saratogaha 32MHz (although its not 100% correct)
01:26:33LloreanAs compared to?
01:26:57Beta2K209MHz on the Gigabeat?
01:27:11 Join hardcore [0] (
01:28:04saratoganormal was 36MHz or so
01:28:25saratogaplus I think I saved 16KB of IRAM
01:31:12 Join webguest69 [0] (n=4aecb47c@gateway/web/cgi-irc/
01:31:12 Quit fyrestorm (Read error: 104 (Connection reset by peer))
01:31:20webguest69hey anyone there
01:32:03webguest69stupid noob question can i put rockbox on a 6th gen classic
01:32:13webguest69through any unoffical or offical means
01:32:19Beta2KNope :)
01:32:34webguest69anyhacks at all then for it
01:32:38Beta2K6gen is toally different hardware then any of the previous players
01:32:40stripwax no
01:32:45webguest69i know thta
01:32:55delorean90you could try porting it yourself
01:33:02Beta2KAnd the FW is encrypted
01:33:29webguest69then we must unencrypt it .. but I cant do that I Hack games not ipods
01:33:47stripwaxwell, yes
01:34:15webguest69ill be back soon
01:34:49 Quit obo ("bye")
01:35:03webguest69hey are the forums down?
01:35:12webguest69coldnt login earlier
01:35:30webguest69then wtf y cant i login
01:35:37stripwaxwhat error does it give you?
01:35:42stripwax(we can't read minds unfortunately)
01:36:01webguest69it loads for a hour and then cant display page
01:36:08stripwaxfor an hour?
01:36:17stripwaxtry again?
01:36:18webguest69(bye hour I mean 5 minutes or so)
01:36:26stripwaxwffm, right now
01:36:37LloreanI'm having no problem at all with the forums right now
01:36:41webguest69wtf does wffm mean
01:36:46stripwaxworks fine for me
01:36:54webguest69ahh okay then
01:37:09webguest69well ... i dont know
01:37:12Lloreanstripwax: Please, use full words. I assumed that was just some sort of typo, and you know the channel guidelines.
01:37:25saratogainterestingly, only half the speed up from the Tremor IMDCT is due to ASM, the rest is just a better algorithm
01:37:41Lloreansaratoga: So we can maybe eke some more speed out of it?
01:37:47stripwaxLlorean - sorry-
01:37:56 Join fyrestorm [0] (
01:38:30saratogasure i think WMA can be made a lot faster
01:39:51 Quit goibhniu (Read error: 110 (Connection timed out))
01:39:53LloreanMaybe AAC too?
01:43:24delorean90would this be the proper place to discuss new ports? or should i stick to rockbox-community?
01:43:39LloreanCommunity is for non-Rockbox stuff.
01:44:04webguest69whats the new port
01:44:29saratogaLlorean: definately AAC
01:44:29delorean90its not a port yet, but the sansa view
01:44:44saratogaAAC-LC and WMA are practically the same thing
01:46:35saratogathe View is an interesting target because of the Nvidia-PP chipset
01:46:37webguest69anywere i can buy an ipod video localy you think
01:46:49webguest69i mean what kind of store would have one
01:47:07saratogahowever, unless you can get a team of very interested hackers, or a datasheet from Nvidia, I think it would be a very difficult port
01:47:17webguest69Becasue i need a new ipod by midnight and it has to be hacable
01:47:39saratogathe PP targets, on which the chipset used in the View is apparently based, have been under development for many years now and are only just now becoming well supported
01:48:09Lloreanwebguest69: The only stores that are likely to have one are pawn-shops and non-chains.
01:48:15webguest69hate to be selfcentered but i am freaking out my ipod broke and i am gona be gone for a month and need one
01:49:14 Join dataangel [0] (
01:50:07dataangelWhat's the difference between plugin_get_buffer and malloc? Do you use it like you would malloc?
01:50:28stripwaxit's a static buffer
01:50:36saratogai think plugin_get_buffer just gives you a pointer to the entire audio buffer
01:50:41bluebrotherthere is no malloc in Rockbox, so there is no difference ;-)
01:52:47dataangelis the app in charge of keeping track of how much of the buffer its used?
01:53:03dataangelthe comment says it returns how much is left... do plugins get a set amount to start or ... ?
01:53:23saratogathe amount available depends on how much ram you have
01:53:58dataangelwhat determines how much a plugin gets though?
01:54:00*dataangel digs around plugin.c
01:54:18LloreanIt's kinda all or none.
01:54:31LloreanYou either don't use the audio buffer, or you stop playback and use all of it.
01:54:53bluebrotherwell, you don't need to use all of it. You get all :)
01:55:18dataangelmy app needs to play audio anyway
01:55:31bluebrotherthen you can't get the audio buffer.
01:56:01LloreanDepends on what he means.
01:56:16LloreanIf he's playing audio within the app itself, as in the app is doing the playing, the buffer's his to do with as he pleases.
01:56:16dataangelit plays like 3 second sound files every few seconds
01:56:44bluebrotherLlorean: right. I assumed playing back means using the playback engine.
01:57:14dataangelhrm, actually there's plugin_get_buffer and plugin_get_audio_buffer
01:58:25 Quit webguest69 ("CGI:IRC (EOF)")
01:58:27*Llorean is not at all familiar with the specific function names.
01:58:41LloreanI would hazard that one gets space from the plugin buffer, and the other from the audio buffer, then?
01:58:46dataangelso plugin_get_buffer gets some amount specific to yer player, determined by PLUGIN_BUFFER_SIZE, and then there's the separate audio buffer
01:59:23dataangelat a glance I don't see a way to tell the api that you're using any of the plugin buffer though, which is why I'm curious if you manage it yourself
02:00:09*dataangel should just shutup and read more source code ;p
02:00:09Lloreandataangel: Tell it you're using it for when you do what?
02:00:27dataangelLlorean: So if you call it again later with a different size it can determine whether anything is left
02:00:55dataangelerr, I see size parameter is to request, not to find out how big what you got back is
02:02:00dataangelerr, no it's just what's left if whole buffer... meh need to read
02:04:19Lloreanbluebrother: I know it used to be outside the folder, I remember when it changed. I just meant that he needs to be looking in it since he's using rbutil. :)
02:04:42LloreanHe's updating the bootloader anyway, or trying to.
02:05:23saratogaok sounds like I fixed the audio skipping problem
02:05:52saratoga(forgot to convert between the vorbis and wma fixed point formats)
02:07:12bluebrotherLlorean: I'm sure you're aware of that. Just wanted to point out that there was a change (as he said he hasn't updated in like 4 years)
02:07:41bluebrother4 years ... that needs to be sometime in the early stages of the h100 port. Wonder why he never updated in between.
02:07:43stripwaxlame_128 - 209% realtime, vorbis_128 - 253% realtime. curiously, above 96kbps, mp3 doesn't seem to scale so much with bitrate unlike vorbis (lame_256 is 199.8% realtime, vorbis_256 is 215%)
02:08:41saratogastripwax: I've been wondering about that too
02:08:44bluebrotheranyway, time for sleep now, it's quite late here. Just wanted to finish the translation update, but unfortunately the previous lupdate gave me a nice bunch of conflicts :(
02:08:54 Quit bluebrother ("sleep. now.")
02:08:54saratogatheres no good reason for that to happen
02:09:39saratogastripwax: are you interested in codec optimization?
02:10:31stripwaxI'm interested in performance per watt
02:11:17saratogathe synth + DCT functions in synth.c use up ~50% of the entire codec runtime for 192k mp3, and they should be completely invariant with changing bitrate
02:11:28*mrkiko restarts it's brain
02:11:43saratogayou could try commenting them out and benching that 96k file again
02:11:53 Quit ender` (" Intelligence is the ability to avoid doing work, yet getting the work done. -- Linus Torvalds")
02:12:11saratogamight be interesting to see where all that improvement in CPU time really comes from
02:13:36saratogaanyone object to me leaving the old WMA IMDCT in but ifdefed out in case anyone ever wants it again?
02:14:01Lloreansaratoga: It'll be in the SVN history anyway, right?
02:14:08stripwaxnot sure I understand - comment out what part exactly? or does 96kbps not need synth and dct?
02:14:36 Quit culture (Read error: 110 (Connection timed out))
02:14:38saratogastripwax: if you comment out parts of the decoder and benchmark, you can see how much time the codec spent in those functions
02:15:01saratogaLLorean: thats a good point
02:16:00saratogaexample: codec runs at 200% with everything enabled, commenting out synthfull and dct32 raises you to 400% realtime, this means that the codec spent 50% of its time in those two functions
02:16:13stripwaxsaratoga - that would only work for the parts after bitstream but I see your point. Would be better to get the profiling build working though
02:16:32saratogastripwax: it should work for any part of the codec
02:16:57saratogayou can decode a blank buffer just as easily as a full one
02:17:30stripwaxbut then you're decoding a blank buffer and/or just testing error paths on garbage buffers .. that won't help you determine anything
02:17:59stripwaxA profile build should be easier
02:18:27saratogaa profile build would be easier if we knew how to make one
02:19:25 Join KimWoojin [0] (n=79b459ce@gateway/web/cgi-irc/
02:21:03stripwaxsaratoga - It worked for me fairly recently
02:23:39 Join ellomates [0] (n=4aecb47c@gateway/web/cgi-irc/
02:23:46stripwaxAlthough checking irc logs, "fairly recently" might be about this time last year..
02:24:03 Quit KimWoojin (Client Quit)
02:24:03ellomatesis there rockbox for the new ipod nano
02:24:09Lloreanellomates: No.
02:24:11ellomateslet me guess no
02:24:27ellomatesabsoultly no way?
02:24:36ellomatesbesides making my own and all that
02:24:40LloreanAbsolutely nobody who wants it has actually done the work.
02:24:42stripwaxyou can't even make your own
02:24:57stripwaxwell, you could, but nobody has figured out how yet
02:25:09ellomatesI really would work on it if i was goning to away fora month
02:25:29ellomatessaway fro a month
02:25:36*stripwax doesn't understand
02:25:40ellomates(that sentence was just all wrong )
02:25:40saratogai don't think you really need to let us know that you're not working on it
02:25:49ellomatesyes i do
02:25:58stripwaxeveryone else is also not working on it :)
02:26:06ellomateswell they should tell us that
02:26:25 Join Thundercloud [0] (
02:26:38ellomates... they should tell us that they are not working on it so we dont get our hopes up...
02:26:45saratogaits sort of implicit when you come in here without managing to read the front page that you're not working on anything
02:27:12Lloreanellomates: How about "don't get your hopes up unless someone says they are working on it, and even then, only if they actually have work to show"
02:27:28ellomatesnevermind ....
02:27:53stripwaxellomates - it's mentioned in the forums and wiki also, so we do tell you that nobody is working on it
02:28:27ellomatesno i am not talking about the offical people i am talking about people who are un officail as in me
02:28:42dataangel /join #c
02:29:14ellomatesokay lets stop this conversation becaus eit will end up with me lokking even stupideir (plus i got to go)
02:29:22 Join Strife89 [0] (n=michael@
02:30:03Strife89Could someone please lend a hand with the Pictureflow plugin? It refuses to find my album art.
02:30:35Strife89I've already asked someone, but we didn't get very far:
02:31:04stripwaxellomates - ok. but everyone is sort-of "unofficial", in the sense that anyone can get on and do the work necessary, and most importantly nobody
02:31:53stripwaxellomates - rockbox runs fine on the list of devices mentioned on the front page of
02:32:08*stripwax tries a profile build
02:33:34LloreanStrife89: Does your album art work in the WPS?
02:33:57Strife89I use the /.rockbox/albumart storage method
02:34:23saratogaif arm profiling coudl be made to work, that'd be awesome
02:34:30LloreanI thought it used the same method of retrieving Album Art as the WPS did.
02:34:43Strife89Evidently, not all of them. :(
02:35:01stripwaxsaratoga - I didn't realise it didn't work!
02:35:15LloreanStrife89: Well, I thought it was just one function that was called
02:35:25stripwaxassuming it doesn't. will know later, but my build takes forever
02:36:05Strife89Llorean: I wouldn't know about that; I still can't read much C.
02:37:05stripwaxthe profiling bug I know of is that it will create a profile.out regardless of whether or not profile.out already exists - not looked at the code for that but it seems to create directory multiple entries with the same name, which causes havoc with my laptop OS when I reconnect it. so need to delete profile.out before rerunning a codec
02:37:23 Join babuun [0] (n=hero@unaffiliated/toki)
02:37:30 Quit Toki (Nick collision from services.)
02:37:33LloreanStrife89: Database is initialized?
02:37:44 Join Toki [0] (
02:38:11saratogastripwax: after much effort this winter i managed to get profile.out files created dependably, but the addresses in them didn't seem to make sense to me
02:38:34LloreanStrife89: And you're using small enough bitmaps, and you've stopped playback?
02:38:35saratogaafter some effort i just gave up trying to correlate them to symbols in objdump or map files, though maybe i just didn't know what i was doing
02:39:04Strife89Yes and yes. :(
02:39:09 Part babuun
02:39:15LloreanStrife89: Completely unpatched build?
02:39:19 Quit Toki (Nick collision from services.)
02:39:33 Join Toki_ [0] (
02:39:44Strife89Yes (although it's r17883)
02:39:59*Llorean shrugs
02:40:04Strife89I haven't seen any Pictureflow updates, though.
02:40:34Strife89I'll try what's mentioned here (just found it):
02:40:36 Quit saratoga ("CGI:IRC (EOF)")
02:42:46Strife89Nope. No luck.
02:43:23***Saving seen data "./dancer.seen"
02:44:18Strife89Actually, I have three albums, each in their own folder, with album art in the same folder as ./cover.bmp. Those three - and only those three - show up in Pictureflow.
02:45:23Strife89So it definitely seems that /.rockbox/albumart/artist-album.bmp is not implemented in Pictureflow.
02:49:08 Quit dataangel (Read error: 110 (Connection timed out))
02:49:49Strife89I have no idea if this is it or not, but this part of the Pictureflow code seems relevant.
02:50:36Strife89{For log readers: As of August 5th, 2008, the linked pastebin will be gone.}
02:50:38 Join m0f0x [0] (
02:50:46 Quit faemir (Client Quit)
03:01:40 Quit hardcore (Read error: 110 (Connection timed out))
03:02:14 Join dataangel [0] (n=prophet@
03:02:27 Quit herrwaldo (Remote closed the connection)
03:03:23 Quit EspeonEefi ("さよなら")
03:06:32 Quit ompaul (Client Quit)
03:08:48 Join hardcore [0] (
03:10:17 Quit CaptainKewl ("( :: NoNameScript 4.02 :: )")
03:14:33 Quit tvelocity (Remote closed the connection)
03:15:33 Quit hardcore (Read error: 60 (Operation timed out))
03:15:56 Part Strife89
03:20:42 Quit fyrestorm (Read error: 113 (No route to host))
03:21:00 Quit Thundercloud (Remote closed the connection)
03:22:26dataangelWhy does set_color take a title parameter?
03:22:58 Join fyrestorm [0] (
03:23:57 Join Thundercloud [0] (
03:26:53 Join hardcore [0] (
03:35:01 Join Vyse [0] (n=42292add@
03:35:23 Quit DerDome (Nick collision from services.)
03:35:24 Join DerDom1 [0] (
03:35:36 Nick DerDom1 is now known as DerDome (
03:35:37 Quit fyrestorm (Connection reset by peer)
03:36:30 Join fyrestorm [0] (
03:38:01 Quit Vyse (Client Quit)
03:38:04 Join VyseZero [0] (n=42292add@gateway/web/cgi-irc/
03:42:00 Quit VyseZero (Client Quit)
03:54:30 Quit hardcore (Read error: 110 (Connection timed out))
03:56:04 Join Zarggg [0] (
03:57:07 Join hardcore [0] (
03:57:33soapLlorean, JdGordon is not here to defend himself, so I in no way wish to attempt to write the definitive scripture, but I am curious as to what your opinion is regarding my decision to limit the sharing of "corrected" .wps files on the forums. As I am sure you are well aware the WPSs in question were ones originally posted by PHK, and had no expressed license attached, thereby (to the best of my understanding) being All Rights Reserved. JdGordon thought
03:57:33soap I went too far in my clamping action, and while my heart agrees with him my head is having trouble.
03:57:49 Quit Thundercloud (Remote closed the connection)
03:58:09*JdGordon is here
03:58:22JdGordoni didnt say you went to far... just that I think its rediculous
03:58:51Lloreansoap: Rationally, I don't think we can be lenient on this one. If a theme isn't licensed as modifiable, people can't post updated versions.
03:59:10LloreanIf you start acting like there's a gray area, people are going to argue about it to no end regarding other things.
03:59:24soapok, then, JdGordon - perhaps I misunderstood you. I thought you called my actions ridiculous.
03:59:27LloreanIt's one of the few parts of enforcement we can actually be objective on.
03:59:53Ice``Llorean: VLC is illegal in france because of the law I told you about...
04:00:02JdGordonsoap: no, I meant copywrite on that text...
04:00:15Ice``I think that it will be the same with rockbox.
04:00:21LloreanIce``: VLC can use decss to playback copy-protected DVDs, among other things, I assume.
04:00:29JdGordonit was just in reference to your message
04:01:07soapYou think it is ridiculous to apply copyright to the .wps file?
04:01:51Ice``Llorean: yes, it can. But it's just a question of time before rockbox will be declared illegal.
04:02:12Ice``Rockbox is done by reversing firmwares, and, THIS, is illegal.
04:02:31soapwhat about /firmware/events.c
04:02:33LloreanIce``: Actually, a large portion of Rockbox is done by using freely available datasheets
04:02:43LloreanIt doesn't require in any way the presence of the original firmware on the device.
04:02:46Ice``Large != entire.
04:02:56soapignore the header and it is shorter than many .wps files, and arguably contains less creative content.
04:03:03JdGordonIce``: so when that happens, we'll stop distrobuting in france, and have a massive effort to covertly get the fe french devs out of the country :)
04:03:12LloreanIce``: Depends on the player. But if this is true, Linux is illegal in France too, for the same reason.
04:03:38zeLlorean: why linux?
04:03:39Ice``Llorean: You know, these technical arguments may be right, the juge could condamn rockbox anyway.
04:03:43LloreanJdGordon: WPS files fall under copyright protection in the same way bitmap images, lyrics, or any creative work.
04:03:53*gevaerts wants to know since when reverse engineering is illegal
04:04:04LloreanIce``: There's nothing you can do to stop that, and no reason to worry about it since there's nothing to be done about it then.
04:04:07Ice``JdGordon: I'll do my best to get out of that shithole.
04:04:11JdGordonsoap: yes, I see your point.. but imo its silly
04:04:14Lloreanze: Plenty of drivers in it are reverse engineered.
04:04:22Ice``Llorean: Linux will be declared illegal too.
04:04:25zeLlorean: really? why do so many go through clean room then
04:04:39Ice``VLC already is illegal, even the source code is
04:04:41LloreanJdGordon: Unfortunately, silly or not, we need to kinda be all-or-nothing in terms of copyright on the forums
04:04:52Lloreanze: "Clean Room" isn't valid in all jurisdictions.
04:04:59zecrappy :/
04:05:02LloreanIce``: I don't care about VLC.
04:05:03Ice``A friend of mine is working for gsoc on vlc player
04:05:09LloreanVLC doesn't matter here, really
04:05:17Ice``and he risks 300k E and prison to work on it.
04:05:46soapJdGordon, I'll agree that slapping someone for posting a helpful fix to an abandoned .wps file feels stupid as tits on a bull, and strongly feel it is a perfect example of the unreasonable results which can occur by taking only reasonable steps - BUT all that being said I don't believe it is in anybody's best interest to do otherwise.
04:06:05Ice``Llorean: I understood that you are "rigid" with the chan subject and I respect this. But it was to illustrate what I said.
04:06:31zewhat about a patch?
04:06:48soapTHOUGH (and Ll will probably have more problems with this than most) there is a strong case to be made for respecting the ideal of "reasonable" and ignoring a case exactly like this because they are reasonable.
04:06:59soaps/they are/it is/
04:07:06zedoesn't break copyright mere to distribution instructions for someone to mod it themselves (no distribution of the copyrighted code) does it?
04:07:31Ice``In no time, frenchmen will go in jail for breakin' a device, authorities pretexting "reverse engeenering" at any clumsiness...
04:07:33Lloreansoap: My problem with that is that, while I definitely agree these fixes are reasonable, we can't leave copyright up to interpretation without leaving ourselves open to other arguments later.
04:08:58soapMy solution is simple (if not too late): pound the idea into people's heads that they need to expressly state a license from day one.
04:09:08JdGordonthen we shuold force all themes to have redistrobution rights on everything, and modification rights on all the text (better would be just force a CC licence on everything)
04:09:15LloreanAnd our new theme site will require a license that allows modification (and thus fixing) anyway, I believe.
04:09:33LloreanI though a CC-by-SA was planned as a requirement for themes on the new site anyway
04:09:37soapI agree 100% - do we have /any/ right to retroactively force a license:?
04:09:49LloreanI wish we did.
04:10:06LloreanIf we had a license on the wiki, all themes would already be licensed.
04:11:28zeeven if you just considered themes as part of rockbox code base they'd at least be automatically GPL wouldn'tthey? :p
04:11:39soapMarcGuary has been waist-deep in the wiki for months now. Perhaps we should lock him in a room and have him write a replacement wiki from scratch and license it properly. ;)
04:11:41zeagain, too late and unoptimal for this point
04:12:14Lloreanze: Can't really do that though. They're not derivative works, otherwise, for example, bitmaps would be derivative works of Windows Paint or whatever.
04:12:22Lloreansoap: Good plan!
04:12:36zeLlorean: well, they use functions provided by the code..
04:12:48zeLlorean: its more akin to use of a library than a program
04:12:50LloreanNo, they don't
04:13:02LloreanThey're a text file interpreted by the code.
04:13:11LloreanOtherwise bash scripts would fall under the same license as bash itself.
04:14:39zeyeah i suppose the copyright on the particular design arrangement trumps the functional tokens anyway
04:14:51zein a sense
04:15:55Lloreansoap, JdGordon: fortunately, this whole thing should be resolved with the new theme site, where we can not only force a license, but since we have, script updates to themes when possible
04:16:20JdGordonyeah, but that site could still be 2 years away...
04:17:26soapwith the recent commits, how much more is the WPS syntax expected to change in the next 12 months? Multifont is possibly on the horizon, but what else?
04:17:32zewhats wrong with suitable stopgap updates for the current site if the new ones gonna take long?
04:17:44JdGordonmultifont wont break compatibility
04:17:51JdGordonbut nothing else is "Extected"
04:17:56JdGordonexpected even
04:18:28JdGordonbut then again.. had you asked that 2 months ago you'd have got the same answer
04:18:33soapwell, I know unexpected things can come up - but I'm not sure what WPS ideas are in place now which weren't at least discussed 12 months prior to implementation.
04:18:48 Quit dabujo (Read error: 110 (Connection timed out))
04:18:49soap(and I'm sure you do)
04:20:10JdGordonalso, its not always clear if a change being discussed will need to break anything... %p,%P were only changed because it turned out we had to, not because we wanted to
04:20:31JdGordonbut yeah, the syntax should be stable for the next while now
04:21:42*JdGordon goes to have breakfast, and then... continue in his search to make DB playlists resumable
04:22:37soapI thought you were going to be punitive ;)
04:23:32zeLlorean: so if some random stranger reverses something and writes a doc, and then i wrote a driver based on info in the doc, my driver would be illegal in france anyway?
04:23:57 Quit jhulst_ (Read error: 60 (Operation timed out))
04:24:07Lloreanze: Dunno, I'm not familiar with the law
04:24:25zeLlorean: on the other hand i suppose companies would get away with anyway because they'll do all their reversing in secret :p
04:24:34zeso its just opensource that suffers
04:26:20zeso what if nobody fessed up to reversing anything and there wasn't a way to prove something was based on knowledge gained from a reverse engineer...
04:26:37zehow could they enforce it?
04:26:40Lloreanze: When has a lack of evidence stopped a determined law suit?
04:27:18 Join goffa_ [0] (n=goffa@
04:27:26soapHorribly off topic now, but proof is irrelevant. All you need is the law banning it so you have pretense to strategically sue.
04:27:26zewell if you're just gonna aknowledge the system's fundamentally broken and wrong, then whats anything matter, it all becomes arbitrary and meaningless
04:27:58zebut yeah sorry, was just curious... consider it dropped by me
04:29:17ellomatesI just bought a ipod classic (74.6gb)
04:29:23 Quit fdinel (Read error: 110 (Connection timed out))
04:29:32soapnobody working on it :(
04:29:38ellomatesI know
04:29:51ellomatesWait were you talking to me
04:30:25ellomatesokay ill take that as a yes
04:30:30soapyes - I was trying to smartassedly answer your question regarding Rockbox and the Classic. ;)
04:31:02ellomatesanyway how the hell do these things get cracked i mean what do they do to get in there with the code
04:31:22ellomatesI know how to code game hacks but ipods just confound me
04:31:22LloreanDiffers from player to player.
04:31:39LloreanNobody can really provide you with step by step instructions, it's something that requires someone clever to figure all the right things ous
04:31:51ellomatesdiffers from player brand to player brand or indivudal player to player
04:32:20LloreanIndividual player
04:32:25ellomatesor is it there are big diffrences between brands and smaller difrences between players
04:32:34soapa game hack is easy in that you can modify the binary and your PC will happily run the modified binary. The iPod Classic will most likely not run modified code.
04:32:52ellomatesyes it is soap yes it is
04:33:09ellomatesyah i doubt it will too
04:33:20ellomatesor they would have some more work done bye know
04:33:55ellomateslike how do they even begin seeing code
04:34:55ellomatesdont you love pirating stuff and stealing from all companies using google
04:35:11ellomatessorry that was just something i type to see if my chat is monitered
04:35:12soapno, and Rockbox has nothing to do with pirating
04:35:17ellomatesi know
04:35:21Lloreanellomates: Again something else they have to figure out: Where is the firmware on the player, and how do I get my hands on it?
04:35:32Lloreanellomates: Any more talk of pirating, at all, in any manner, and I'll ban you.
04:35:47ellomatesmabye i f we took it apart and went straight at the hardrive
04:35:54LloreanWe do actually respect copyright here.
04:36:10ellomatesdoes anyone know were i can find a gallery of the insides of one a classic
04:36:21ellomatesI respect copyright
04:36:45ellomatesI just sayed that for .. well no real reason but... well does it really matter why...
04:37:02ellomatesbtw a gallery insides
04:37:50Lloreanellomates: Please don't do that.
04:38:03LloreanIf someone has the answer to your question, they'll answer it. You don't need to repeat it over a short period.
04:38:24LloreanFor really basic things such as this, though, you should probably spend some time with Google, or see if anything has been posted to the Rockbox wiki
04:38:36 Join AndyI [0] (n=pasha_in@
04:39:06Ice``Or, you can repeat it, post it in a bunch of threads, write it in CAPS LOCK, in red, and seek for Comic Sans MS font; and then, wait for the ban.
04:39:16 Quit goffa (Read error: 110 (Connection timed out))
04:39:42Ice``(Oh, I forgot the bold, and the underline statments)
04:40:16ellomatesI like ices idea i think i wll do it
04:40:29ellomatesLets put it on apples forum too
04:40:40Ice``(sorry for the caps lock, then)
04:40:51Ice``ellomates: try msdn
04:40:51ellomatesall the forums ill make a script to re post a new post every 5 minutes
04:40:54LloreanTry to stay on topic.
04:41:02ellomatessorry got caried away
04:41:05Ice``Rah sorry
04:41:14Ice``my bad
04:41:39ellomatesmy bad too D :
04:42:15Ice``So, we're sorry, and I'll do an effort to (stay on topic || be banned).
04:42:37Ice``hem, || stands for XOR, ok?! =$
04:43:24***Saving seen data "./dancer.seen"
04:45:53 Quit Zarggg ()
04:46:53ellomatesone off topic question since things have died done for a bit? can anyone sugjest some good bands i like :simple plan,Linkin,park, blink182, bowling for soup and many other punk rock bands.... to stay on topic doesnt the ipod classic run on a harddrive?
04:47:37Lloreanellomates: No off topic questions, please. The channel is logged, and we like to keep it from being filled with unrelated chatter.
04:48:06LloreanAnd yes, the Classic uses a HD.
04:49:45 Quit AndyIL (Read error: 110 (Connection timed out))
04:50:38 Join miepchen^schlaf_ [0] (
04:51:57 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
04:52:01Ice``HD is the worse idea ever had in a mp3 player design.
04:52:35JdGordonthats nonesence
04:52:49delorean90ice'' why do you refer to || as XOR, i thought it was just OR
04:52:53JdGordonif the first batch didnt use hard disks they would never have taken off
04:53:58Ice``delorean90: cause i'm used to use it as "or" and then I realized than I don't want the && case to happen...
04:54:33delorean90ice'': ohhh ok
04:55:12Ice``that, not that
04:55:23*Ice`` bites his fingers.
04:55:34delorean90ok so to saty on topic, i was looking over the wiki and saw that the articles written about rockbox are uneducated
04:56:16Ice``saty. I'll keep in mide.
04:56:23Ice``raah I go to bed.
04:56:27Ice``Later o/
04:57:15delorean90on that note i will go to bed as well
04:57:41 Quit delorean90 ()
04:58:30 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
05:05:34 Join guest3 [0] (n=4b1634ee@gateway/web/cgi-irc/
05:06:19 Quit guest3 (Client Quit)
05:08:13 Join fdinel [0] (
05:11:40Ice``Good night.
05:13:34 Quit ellomates ("CGI:IRC (Ping timeout)")
05:20:41 Quit hardcore (Read error: 60 (Operation timed out))
05:34:54 Join hardcore [0] (
05:44:46 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:46:06 Join sarixe [0] (n=sarixe@
06:02:16 Join SirFunk_ [0] (
06:18:45 Quit SirFunk (Read error: 113 (No route to host))
06:22:05 Quit hardcore (Read error: 104 (Connection reset by peer))
06:22:12 Join hardcore [0] (
06:25:35 Quit SirFunk_ (Read error: 104 (Connection reset by peer))
06:29:01 Join SirFunk_ [0] (
06:31:18 Join shadoxx [0] (
06:41:17 Join _vertic23 [0] (
06:43:25***Saving seen data "./dancer.seen"
06:54:02 Join EspeonEefi [0] (i=espeonee@CERULEANCITY.MIT.EDU)
06:58:33 Quit vertic23 (Read error: 110 (Connection timed out))
07:00:00shadoxxAnyone have a primer on mailing lists?
07:02:46 Join ctaylorr [0] (
07:12:52 Quit perrikwp (" ajax IRC Client")
07:19:03soapmore than ?
07:20:01 Join jhulst__ [0] (
07:20:01 Quit jhulst_ (No route to host)
07:22:30shadoxxAs in, how do I reply, read properly, etc with a real mail client. Mailing lists confuse me to no end.
07:22:55shadoxxI mean, I get how to subscribe. But as far as following threads, and having them displayed properly
07:25:03soapthat's an issue for your email client. Don't get the digests if you want to follow threads, and any good email client can group by thread for you.
07:25:45soapSame with replying - if you want to participate, not just observe, you really do not want the digest.
07:40:02 Quit hardcore ("Client exiting...")
07:43:23 Quit jhulst__ (Read error: 113 (No route to host))
07:44:16 Join jhulst__ [0] (
07:46:38 Quit ctaylorr (Read error: 110 (Connection timed out))
07:48:04 Quit goffa_ (Read error: 110 (Connection timed out))
08:21:33 Join homielowe [0] (
08:23:11 Quit jhulst__ (Read error: 113 (No route to host))
08:25:28 Quit miepchen^schlaf_ (Remote closed the connection)
08:25:29homieloweCurrently midi file playback on the Gigabeat S is too fast, around 200% approximately.
08:26:07 Join miepchen^schlaf [0] (
08:37:03 Join Rob2222 [0] (
08:43:26***Saving seen data "./dancer.seen"
08:55:17 Quit Rob2223 (Read error: 110 (Connection timed out))
09:01:55*JdGordon hopes its not too early to to shop...
09:02:29JdGordonis there any reason to put a local struct variable which is about 20bytes "static" instead of on the stack?
09:03:04JdGordonits inside a loop which iterates less than 6 times ( i tihnk)
09:06:13 Join d [0] (n=dt@
09:06:41 Nick d is now known as Guest19215 (n=dt@
09:11:46 Join d [0] (n=dt@
09:12:00amiconnhomielowe: Probably the plugin tries to set 22.05kHz sample frequency, but this function doesn't work on the S yet. So it will be output at 44.1kHz, i.e. twice the speed
09:12:11 Nick d is now known as Guest86027 (n=dt@
09:12:33amiconnSince the S does have enough power, it could simply use 44.1kHz
09:12:53homieloweamiconn: So I guess do something similar with Ipods, pp etc,
09:14:00amiconnYes. On ipods we don't have working sample frequency selection either, so the plugin uses 44.1kHz (and struggles with any non-trivial midi file)
09:17:24*amiconn thinks midiplay should be adapted to use core dsp as done for mpeglayer
09:22:40 Quit d-k-t-2 (Read error: 110 (Connection timed out))
09:26:53 Quit Guest19215 (Read error: 104 (Connection reset by peer))
09:27:45 Join homielowe_ [0] (
09:27:45 Quit homielowe (Read error: 104 (Connection reset by peer))
10:02:24 Nick homielowe_ is now known as homielowe (
10:07:18 Join moos [0] (
10:13:45 Join n1s [0] (n=nils@rockbox/developer/n1s)
10:21:35 Join KimWoojin [0] (n=79b459ce@gateway/web/cgi-irc/
10:23:02 Quit KimWoojin (Client Quit)
10:24:23 Join stoffel [0] (
10:27:02 Join lolmaus [0] (i=lolmaus@
10:27:05n1shmm, are the wheel directions on the ipod simulator backgrounds reversed?
10:28:50n1sshouldn't turning the wheel clockwise scroll down/go forward?
10:43:27***Saving seen data "./dancer.seen"
10:51:36 Join ender` [0] (
10:52:25 Join hannesd [0] (
10:55:13 Join ZincAlloy [0] (n=d9eed80d@gateway/web/cgi-irc/
10:55:45 Quit lolmaus ()
11:01:04 Join merbanan [0] (n=banan@
11:05:20 Quit stoffel ("leaving")
11:05:43 Quit ZincAlloy ("CGI:IRC (Ping timeout)")
11:07:29 Quit DerDome ("Leaving.")
11:07:45 Join DerDome [0] (
11:08:08JdGordondoes anyone have an email addy for domonoky?
11:19:08 Quit advcomp2019 ("Ex-Chat")
11:21:12 Join goibhniu [0] (
11:26:08 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
11:30:05 Quit amiconn (" HydraIRC -> <- Go on, try it!")
11:30:11 Join spiorf [0] (
11:40:21 Join Nico_P [0] (
11:41:50 Join davina [0] (
11:45:28JdGordonSlasheri: hey, you round?
11:47:12 Join d-k-t-2 [0] (n=dt@
11:48:21 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
11:53:35Nico_Pcan someone tell me what the "narrowband" setting in the rbspeex settings of rbutil does?
11:53:46Nico_PI have no idea how to translate it
11:54:42SlasheriJdGordon: morning :)
11:56:33JdGordonSlasheri: hey, (hope im not getting too iritating...) these stuct variable names arnt helping... I tihnk I've sort of figured it out... Is the csi linked to the menus at all? I mean if I change the csi how does the browser know whicih "file" to select?
11:57:21JdGordonIt seems to me that if its changed by something other than tagtree_enter/exit, the browser would show the files properly, but get lost if you try backing out of the curent search?
11:58:33SlasheriJdGordon: yes, each menu entry is linked to a search instruction
11:59:31JdGordonso if the csi is changed, the browser should figure it out safely next reload?
12:00:39Slasheriyep, it should affect the next results from the tagcache query
12:01:08Slashericsi keeps all levels of queries for a single root menu entry
12:01:36 Join desowin [0] (n=desowin@atheme/member/desowin)
12:03:02JdGordonok, now what would happen if a csi was loaded which didnt have a menu assosicated with it? i.e tagnavi changed between boots and the saved csi wasnt reloaded from the config?
12:04:48Slasherihmm, it should be just ignored then if there are no links to that csi entry from menu_entry
12:05:40JdGordonbut it could still generate a playlist from it?
12:06:22Slasheriyep, it should be able to do that just fine
12:06:23*JdGordon wants to dump the csi into .playlist_control as text so playlists could be rebuilt as easily as dirplay
12:06:32Slasherihehe, nice idea
12:07:25 Quit Guest86027 (Read error: 110 (Connection timed out))
12:07:39JdGordonwould it work?
12:09:14Slasherii don't see why it wouldn't, as long as you write some handlers for it in the tagtree
12:09:50JdGordonthat would be much nicer than adding each track individually
12:10:01Slasheriyour "playlist csi" could be for example one static instance of search_instruction which you assign to csi-pointer and then populate the playlist
12:10:05Slasherii think that should work
12:10:29JdGordonits going to be annoying to write the parser for it though :p
12:10:39Slasheriyou mean playlist.c would read the csi?
12:11:05JdGordonno, I'd do it in tagtree to keep it all there
12:11:13Slasheriah, ok :)
12:11:43 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
12:12:29 Join obo [0] (n=obo@rockbox/developer/obo)
12:12:31bluebrotherLlorean: we now have a nice "FAQ" entry in the left-side menu, but the forums page is still lacking it. Can you fix that?
12:13:52JdGordonSlasheri: format_id is the only variable in that struct which actually needs to be "synced" with tagnavi right? but isnt needed to just rebuild the playlist?
12:14:02JdGordonnor is name?
12:15:36 Join ZincAlloy [0] (n=d9eed80d@gateway/web/cgi-irc/
12:16:27 Quit ompaul (Client Quit)
12:16:37SlasheriJdGordon: format_id is only used for sorting (iirc)
12:16:45Slasheriso you could ignore that in the code
12:17:07Slasherior hmm, maybe you couldn't..
12:17:18Slasheriif you don't want the playlist be unsorted
12:18:11JdGordonthat is an index into the saved forus from tagnavi idsnt it?
12:18:24Slasheriof course format_id also defines the format of titles and so on displayed in the browser but for playlist the only function of it is the sorting
12:19:03Slasheriit's an index to the display_format struct
12:19:18JdGordonhmm, might be extra stuff to store then :(
12:19:30JdGordonalthough, how often would tagnavi change?
12:19:38JdGordonprobably not enough to worry about
12:19:59JdGordonalso.. whats the result_seek array?
12:21:09Slasheriiirc, that contains the tagcache index to the last selected object (so that filters can be applied on the next level)
12:21:48JdGordondo I need to worry about it?
12:23:10Slasherihmm, probably that much that you need to populate them for the every level of csi
12:23:35Slasheriand then you can run the final query after you have ran all of the previous level queries
12:23:52 Join mf0102 [0] (n=michi@
12:27:50JdGordonthere isnt a simpler way to do this is therE?
12:28:58Slasherisorry but i don't think so.. please check the retrieve_entries function
12:29:15bluebrotherwhere does the m200 get its firmware file from? Is that stored on the flash memory or a different flash chip?
12:29:54JdGordonyep, I keep getting back to that.
12:30:09JdGordonnot quite sure about rebuilding that result_Seek array
12:30:55Slasheribasically that retrieve_entries functions does all the work of parsing the csi and storing those values
12:31:17Slasherieven when adding tracks to the playlist, results are retrieved by using the retrieve_entries
12:32:04 Join petur [50] (n=petur@rockbox/developer/petur)
12:32:31 Quit safetydan ("Leaving")
12:41:18 Join herrwaldo [0] (
12:41:36 Quit tessarakt (Read error: 104 (Connection reset by peer))
12:41:50 Join tessarakt [0] (
12:42:37JdGordonSlasheri: ok, I'm now totally confused.. whats the clauses? I thought it was "artist = blaa" "album = bllaa" ?
12:42:53JdGordongoing down the tree though, clause_count stays 0
12:43:30***Saving seen data "./dancer.seen"
12:43:35Slasherino, those are the filters :)
12:43:56Slashericlauses are those expressions made on numeric tags
12:44:34Slasheriand basically filters are built from the result_seek array when advancing to next level
12:45:52Slasheriwell, of course clauses can be done for string tags also.. well, clauses are just more advanced form to filters :)
12:46:31 Quit ender` (" Microsoft is like a case of herpes. You can't kill it off entirely, just force it to go into a dormant stage for a while.")
12:46:41JdGordonnormally, it will only be filters though?
12:47:02Slasheriit depends on the query, what's defined in the tagnavi.config
12:48:10JdGordonsimplest case is just store the filters order and the name of the item instead of the result_seek... then to rebuild just put the filters back in, call retrieve_entries and loop untill we find the result_Seek index for the item we named?
12:50:05 Join PaulJam [0] (
12:50:22Slasherihmm.. you can't really store filters because they are just tagcache index pointers.. but you can store the entire csi and build result_seek thenk
12:52:00JdGordonthe tagorders arnt #defines?
12:52:08JdGordonthe values of the tagorder array i mean
12:52:18Slasheriyes, those are defines
12:52:26Slasheribut not the actual filter (the result_seek)
12:53:13JdGordonok, yeah
12:56:45JdGordonSlasheri: something simple like, ? which assuming tagnavi didnt change, looping through the retrived items should get the right index?
12:57:41JdGordonno.. something must be missing
12:59:31 Quit K4rP4D (Read error: 113 (No route to host))
13:00:31 Quit bluebrother ("bbl")
13:05:14 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
13:06:33 Join culture [0] (
13:08:39 Join alleyoopster [0] (
13:08:50 Quit DataGhost (Nick collision from services.)
13:08:58 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
13:12:56 Join Thundercloud [0] (
13:13:53SlasheriJdGordon: hmm, why not just encode the entire struct from binary to base64 or something like that :)
13:14:09 Part Miles
13:14:54JdGordontest is nicer
13:15:03 Join robin0800 [0] (
13:15:24Slasheribut harder to keep in sync with code
13:15:26JdGordonthat snippet dumps it ok, but the text is the current selection (obviously).. is that enough?
13:16:02Slasherihmm, maybe
13:16:34Slasheriyou are not dumping the clauses?
13:16:58JdGordonnot to start with
13:17:06JdGordonunless I absolutly have to
13:18:26Slasheribtw, iirc the name in csi is completely unused and can be removed
13:19:12 Quit Seed (Nick collision from services.)
13:19:19 Join Seed [0] (
13:25:14 Join [CBR]Unspoken|w [0] (n=cbr@
13:32:00 Join fdinel [0] (
13:32:12 Join pixelma [0] (n=Marianne@rockbox/staff/pixelma)
13:39:02 Quit n1s ()
13:39:28 Join MethoS- [0] (
13:39:54 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
13:45:44 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
13:46:41 Quit Thundercloud (Remote closed the connection)
13:48:12*stripwax looks to see if Llorean is still around
13:50:03 Quit Rob2222 ()
13:51:00 Join Thundercloud [0] (
13:55:18 Join Rob2222 [0] (
14:01:42*amiconn blames linuxstb for breaking the charging screen
14:01:56 Quit spiorf (Remote closed the connection)
14:01:59 Quit culture (Read error: 110 (Connection timed out))
14:05:43 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
14:07:56 Join spiorf [0] (
14:14:04 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
14:16:39 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
14:19:21 Join MethoS_ [0] (
14:22:12 Join thegeek [0] (
14:22:54 Join roolku [0] (
14:25:51 Join stoffel [0] (
14:28:39bluebrotherpixelma: around?
14:29:13 Quit spiorf (Read error: 110 (Connection timed out))
14:30:16 Join Rondom [0] (
14:31:54 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
14:34:00gevaertsJdGordon: did you see FS #9150 ?
14:34:14JdGordonprobably.... remind me of it though?
14:34:43JdGordonah yes
14:34:47 Quit MethoS- (Read error: 113 (No route to host))
14:35:15gevaertsWe really need tool autobuilding
14:35:16JdGordonlet me fix up my /home so my box can do builds again and ill look into it
14:35:21JdGordonand yes we do
14:36:21JdGordonback in a min
14:36:32 Quit JdGordon ("Konversation terminated!")
14:38:00 Join tvelocity [0] (
14:38:30 Join JdGordon [0] (
14:39:07 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
14:40:02 Quit ZincAlloy ("CGI:IRC")
14:43:31***Saving seen data "./dancer.seen"
14:45:04 Join ender` [0] (
14:46:07JdGordongevaerts: how do i build checkwps?
14:47:04Nico_Pbluebrother: I need help to translate "narrowband"
14:47:24 Join alleyoop [0] (
14:47:28stripwaxJdGordon - I think you used to be able to just make from tools, but could be wrong
14:48:15 Join HauDyr [0] (
14:48:43 Quit alleyoopster (Read error: 104 (Connection reset by peer))
14:49:43bluebrotherNico_P: if narrowband is enabled files will get resampled to 8kHz, otherwise files will be 16kHz. See the help text of rbspeexenc
14:49:44stripwaxalthough I don't know what "dw" means, so if it means "don't worry", you can ignore me
14:49:58JdGordonyes :) dont worry
14:50:01bluebrotherapart from that I can't tell you much about that setting ...
14:50:17JdGordoni thought it was make checkwps in tools... turns out there is a checkwps folder :p
14:50:20HauDyrhello i just upgraded to a new build and my wps screen that worked fine on the old build is now broken how can i find out where in the wps file the error is ?
14:50:20Nico_Pbluebrother: where is the help text? you mean using −−help?
14:50:28bluebrotherdw sounds like some stupid hungarian notation variable prefix to me :/
14:50:43bluebrotherNico_P: yes.
14:50:54gevaertsbluebrother: 'double wps' ?
14:50:54 Quit petur ("later")
14:51:01stripwaxif i'm not allowed to say "wffm" in this channel, I think "dw" should definitely go too.. :)
14:51:08bluebrother"rbspeexenc expects a mono 16 bit WAV file as input. Files will be resampled to either 16 kHz by default, or 8 kHz if narrowband mode is enabled."
14:51:21bluebrotherwffm? What's that?
14:51:26stripwaxHauDyr - you can use "checkwps"!
14:51:39stripwaxbluebrother - "works fine for me", which I thought was in pretty univeral usage
14:51:41bluebrothercheckwps is currently broken ...
14:51:50HauDyrbluebrother, heh
14:51:51pixelmabluebrother: now I am
14:52:22bluebrotherpixelma: nice. I'm wondering how to translate "build" to german. That's mostly the unfinished translations in rbutil
14:52:28HauDyrbluebrother, then how do i fix checkwps :-)
14:52:35*JdGordon is fixing it
14:52:51bluebrotherapart from that (as I reworked it rather completely) I'd be interested if you think the current way is good or not
14:54:11pixelmabluebrother: good question, no idea. I usually talk about builds myself, even in German... :\
14:55:29bluebrotherhmm. Sounds somewhat "user-incompatible" to me.
14:56:29pixelmaI know... just translate it as "Version" ?
14:56:34gevaertsbluebrother: Gebäude ?
14:56:35*gevaerts hides
14:56:49pixelma(of the software...)
14:56:50 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
14:57:06bluebrotherhow do you translate "build version" then? Versions-Version? ;-)
14:57:18pixelmagevaerts: wouldn't that be "building"?
14:58:42bluebrotherwhy don't we have a swedish translation for rbutil btw?
14:59:03gevaertspixelma: possibly. I think that "build" as a noun is not very clean anyway
15:00:52HauDyrwhile i am on the channel i just want to thank all of you for building the great piece of soft/firm-ware that RockBox is, my sansa would be a POS without it.
15:02:33bluebrotherPOS == Point of Sales? ;-)
15:02:44HauDyrbluebrother, excatly :-)'
15:04:19pixelmabluebrother: seems German wikipedia is a bit denglisch here too -
15:04:43 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
15:04:55gevaertsHauDyr: a simulator should also give you the info you need
15:09:48 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
15:14:20 Join Arathis [0] (
15:14:31 Join d [0] (n=dt@
15:14:47 Nick d is now known as Guest41370 (n=dt@
15:17:32 Join d [0] (n=dt@
15:17:57 Nick d is now known as Guest37708 (n=dt@
15:21:19 Join wolle- [0] (
15:21:41wolle-Hello, can you say me when rockbox will be able to work on e250 v2?
15:21:51Ice``some day.
15:22:29PaulJamor never...
15:23:06wolle-still working on it...
15:23:06wolle-or not..?
15:23:15*bluebrother points to the wiki: "we do not estimate timeframes".
15:23:37bluebrothercheck the New Ports forums. All progress is there.
15:23:40PaulJamwolle-: there is a thread in the new ports section in the forums where you can read about the progress.
15:23:44Ice`` me when rockbox will be able to work on e250
15:23:46Ice`` v2?
15:24:02Ice``(sorry for whitespaces, it's because of irssi)
15:24:34*bluebrother wonders what's up with Ice``
15:24:47Ice``He said
15:24:56Ice``Hello, can you say me when rockbox will be able to work on e250 v2?
15:25:02 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
15:25:08bluebrotherand why you're repeating it? Even in case, this channel is logged
15:25:08*gevaerts saw that as well
15:25:24Ice``so I said "me when rockbox will be able to work on e250 v2?"
15:25:37*Ice`` superhides.
15:25:53*bluebrother doesn't understand the use of that
15:26:05wolle-oh let it better
15:26:10wolle-i think i go better out of here
15:26:39Ice``bluebrother: there is no use, it was just a stupid joke.
15:26:41wolle-i cant find this thread PaulJam
15:26:51wolle-in the new port section with v2?
15:27:27wolle-ah this one
15:27:28wolle-thank you
15:27:54bluebrotherah. Didn't got that.
15:28:20wolle-thanks thats all i need
15:28:23bluebrotherand btw, s/stupid/annoying/
15:28:28wolle-have a nice day PaulJam and bluebrother
15:28:33 Quit wolle- ()
15:29:23Ice``bluebrother: we annoy each other, but this time I didn't brake with chan rules...
15:29:33Ice``There isn't any rule about jokes.
15:30:02Ice``at least, I hope.
15:30:49 Quit dataangel (Read error: 110 (Connection timed out))
15:32:33 Part Ice`` ("So. There is a lot of more welcoming chans on this network. I'm sorry for your skimpy minds, but afterall, that's not my prob)
15:35:48JdGordonNico_P: 8771 should be fixed yeah?
15:36:48 Quit d-k-t-2 (Read error: 110 (Connection timed out))
15:38:12Nico_Pyeah, should be
15:38:13*amiconn ponders making the usage of a backdrop (on greyscale & colour lcds) part of the drawmode
15:39:16Nico_PJdGordon: I think it should be marked as related to the real bug before being closed
15:39:18amiconnThis would simplify the drawing code, and would also allow per-viewport selection of backdrop usage (if one is set)
15:39:34pixelmabluebrother: got a manual question. There is currently a "Getting Extras" paragraph which only talks about where to download the fonts package which of course doesn't apply to the charcell Player, and its manual currently looks like this:
15:40:32pixelmashould the whole "Getting Extras" part be disabled there or a sentence that there are no extras available - or even something else?
15:40:56JdGordonNico_P: which task was the real bug?
15:41:05Nico_PI added it
15:41:33JdGordonah ok, im too slow :)
15:41:48JdGordonamiconn: yeah, that would be good.
15:42:36*gevaerts thinks the ideas in could be nice
15:42:51bluebrotherpixelma: looks nice :) Well, I'm not 100% sure −− following the way the manual is done the paragraph should get removed. OTOH it might be helpful for new users if there it's explicitly mentioned that there are no extras for those player.
15:42:52amiconnThe backdrop would still cover the whole screen, but some viewports would replace it by the bg colour in their area
15:43:01Nico_PJdGordon: I closed the task
15:43:15bluebrotherbut I think I prefer removing the whole paragraph
15:44:35amiconnJdGordon: Problem is that if I change it that way, backdrops would disappear until all other code is adapted
15:44:54amiconnI can prepare it though, so that the drawmode always includes backdrop usage if one is set
15:45:09 Quit Guest41370 (Connection timed out)
15:45:32JdGordonisnt there a STYLE_DEFAULT mode? just set that to show the backdrop?
15:45:47JdGordonwhich is a ensible default anyway
15:46:04amiconnThat one isn't drawmode related at all
15:47:50pixelmabluebrother: I also think that the introduction which is there for all the other manuals is too font "centric", it should rather be named "Getting Fonts"... :\
15:49:52bluebrotherthere's the note about some themes requiring fonts. That could get scrapped completely from the player manual
15:50:10JdGordonwe really need a player manual? :p
15:50:59bluebrotherno. Lets trash all manuals. Who needs a manual anyway? ;-)
15:51:24*JdGordon wont complain
15:51:38bluebrotherthat's for sure :P
15:53:09PaulJambut would you also volunteer to adress all the support requests?
15:54:43 Part goibhniu
15:55:00*JdGordon has to do everytinh!? :p
15:55:05 Join MethoS- [0] (
15:58:18 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
16:04:30 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
16:05:38 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
16:05:41 Quit advcomp2019 (Client Quit)
16:06:17 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
16:09:14 Join stripwax [0] (
16:12:22 Quit MethoS_ (Read error: 110 (Connection timed out))
16:14:40PaulJamhmm, i still get errors when trying to compile checkwps on cygwin.
16:16:46JdGordonmr500 and d2 though yeah?
16:17:40HauDyri get a segfault when i try to use the sansa version
16:18:44PaulJamJdGordon: i get this error for every target:
16:19:25JdGordonwierd... clean builds here
16:21:51JdGordon-g is enough for gdb isnt it?
16:25:40JdGordonoh great.... segfault and gdb isnt being helpful
16:25:43 Quit robin0800 (Read error: 104 (Connection reset by peer))
16:26:25HauDyrtry backtrace
16:28:30 Join robin0800 [0] (
16:29:09HauDyrmine says that there is problems with the getwith() in screen struct
16:29:16JdGordonyep, fixing
16:29:27HauDyrsuper duper
16:30:52*gevaerts has been wondering about the tracker
16:31:09gevaertsWasn't there going to be a dev-only feature request tracker ?
16:31:38JdGordonZagor said it couldnt be done in the one project.. so we are using the forums
16:32:06LloreanJdGordon: No, we were gonna use the forums anyway
16:32:28Lloreangevaerts: There's no way to split access by groups for flyspray, so we can't have one until we come up with a new solution.
16:32:45JdGordonHauDyr: fixed
16:33:17LloreanJdGordon: The idea was that people can talk about features in the forums/list/whatever, and devs who like one can post it to the dev-only feature request tracker
16:33:18gevaertsok. I must have missed that information somehow
16:34:07*gevaerts thinks we need flyspray for that
16:34:26 Join MethoS_ [0] (
16:36:43 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
16:36:47 Quit _vertic23 (Read error: 104 (Connection reset by peer))
16:38:41 Quit Thundercloud (Read error: 104 (Connection reset by peer))
16:39:15 Join vertic23 [0] (
16:43:35***Saving seen data "./dancer.seen"
16:45:40 Quit robin0800 (Read error: 104 (Connection reset by peer))
16:48:40 Quit MethoS- (Connection timed out)
16:49:29 Quit advcomp2019 ("Leaving")
16:52:09 Join MethoS- [0] (n=clemens@
16:52:10 Quit MethoS_ (Read error: 104 (Connection reset by peer))
16:55:29 Join J-23 [0] (
16:55:40J-23Hi! Is there any Backspace equivalent in Rockbox's file renamer?
16:56:22HauDyrJ-23, yes
16:56:27LloreanYou can delete characters. There should be a description of controls for the virtual keyboard in the manual, it varies by target
17:01:20 Join gregzx [0] (
17:12:13 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
17:13:24HauDyris the %pb tag changed? because in cabbiebv2 it is used like this: %pb|pb-176x220x16.bmp|5|182|166|8| in the manual it says that it should be used like this: %pb|height|leftpos|rightpos|toppos| ?
17:14:17pixelmayes, it has changed and the one changing it didn't bother updating the manual, just the CustomWPS in the wiki... ;)
17:15:42 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
17:16:01gevaertspixelma: you chased him away
17:17:50 Join Horschti [0] (
17:18:36 Quit Horscht (Nick collision from services.)
17:20:32 Quit k4y (Nick collision from services.)
17:20:39 Join k4y_ [0] (
17:36:38 Join lolmaus_Melven [0] (i=lolmaus@
17:42:58 Join stripwax [0] (
17:43:44 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
17:44:40 Nick k4y_ is now known as k4y (
17:44:51 Join Acky [0] (
17:46:09 Join robin0800 [0] (
17:54:37 Quit obo ("bye")
17:56:18 Join petur [50] (n=petur@rockbox/developer/petur)
18:00:41 Join MarcGuay [0] (
18:00:54 Join d-k-t-2 [0] (n=dt@
18:02:08 Quit Acksaw (Connection timed out)
18:02:44 Join Vorador [0] (
18:04:05 Quit alleyoop (Remote closed the connection)
18:08:22 Quit freqmod_qu (Remote closed the connection)
18:10:06bluebrotheris our installation of flyspray modified?
18:11:40 Quit stoffel (Read error: 113 (No route to host))
18:13:55 Join quassel251 [0] (n=quassel@2001:700:300:1800:213:d3ff:fee9:5ed0)
18:14:56 Quit quassel251 (Client Quit)
18:15:09 Join freqmod_qu [0] (n=quassel@2001:700:300:1800:213:d3ff:fee9:5ed0)
18:16:34 Quit DarkFox (Connection reset by peer)
18:17:04 Join DarkFox [0] (
18:17:05 Quit nplus (Read error: 104 (Connection reset by peer))
18:19:42 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
18:20:51 Quit Guest37708 (Read error: 110 (Connection timed out))
18:25:55pixelmaMarcGuay: I don't agree with taking out the text editor section from the ManualTodo because I think the description is very sparse. But you could have deleted the line about "wavplay"
18:26:31 Join Zarggg [0] (
18:26:39MarcGuaypixelma: Okay, go nuts.
18:27:09 Join nplus [0] (n=nplus@141.25.Globcom.Net)
18:27:15 Join stoffel [0] (
18:32:50pixelmaMarcGuay: do you think the text editor description is enough?
18:36:47 Join Thundercloud [0] (
18:37:29 Join thebishop [0] (
18:37:49thebishopis there an eta on 6G ipod Classic support?
18:38:29MarcGuaypixelma: The wording is a bit odd, but it says what needs to be said.
18:38:31gevaerts(a) Nobody is working on it, and (b) it's not easy
18:38:39bluebrotherwe don't estimate timeframes and nobody is currently working on it
18:38:59bluebrothercheck the New Ports forums, there is the only progress (if you can call it progress at all)
18:40:14MarcGuaypixelma: And the Line Options Menu AKA the Item Menu (the naming is out of sync?) could use some documenting.
18:41:24pixelmayes, that's my main point - the problem is that there are two menus...
18:42:08pixelmabtw. I think the ManualTodo page is quite outdated itself
18:42:56MarcGuaypixelma: Definitely. I was wondering if it would be wise to purge it and start over...
18:43:13*Llorean sees nothing wrong with purging if it'd be less work to rebuild than to repair.
18:43:28pixelmamaybe it would
18:43:28MarcGuayLlorean: :) Sometimes a tough call.
18:43:38***Saving seen data "./dancer.seen"
18:44:35MarcGuayI'm always a little hestitant to burn the past just in case something gets lost, but if it's important enough we should be able to find it again.
18:44:50LloreanThe wiki keeps revision history.
18:44:59LloreanYou can't burn the past, only bury it for easy recovery. :)
18:45:31markunthebishop: the port might never happen
18:46:03bluebrotherMarcGuay: for some reason the links in your email seem to be truncated ... :(
18:46:25MarcGuayTrue enough. If anyone wants to tackle it, I'm going to spend some time away from the machine, will check later.
18:46:43MarcGuaybluebrother: That's because I'm dumb and copied it from the forum post. Duh.
18:46:48thebishopmarkun, because the ipod firmware has encryption now?
18:47:13*pixelma wonders whether the recent mail activity table on the front page could also have a column showing the mailing list (i.e. -dev or -users)
18:47:30 Quit Nico_P (Remote closed the connection)
18:47:38MarcGuaybluebrother: I'll send a new one out with the right links.
18:47:41markunthebishop: yes, everything's encrypted, and I don't know anyone here working on it.
18:49:30bluebrotherMarcGuay: well, I guess those people interested in doing work on it will manage to fix those links. Btw, I was wondering why you posted to the users-ml. Wouldn't the dev-ml be much better suited?
18:49:50Lloreanbluebrother: I'd say we'd like users to pitch in too, if at all possible.
18:49:55*Llorean thinks it should be on both
18:50:20thebishopmarkun, on older ipods, I though Rockbox wasn't really a firmware update, rather the ipod firmware booted from the partition with rockbox installed... is the difference that the new ipod firmware checks that partition?
18:51:32bluebrotherLlorean: fair point, but right now I'm almost about to unsubscribe from the users ml. Too much noise, and too much whining about etiquette reminders
18:51:44bluebrotherwhich quite starts to piss me
18:52:27 Quit ender` (Read error: 104 (Connection reset by peer))
18:53:17LloreanWe should just shut down the user list!
18:54:24MarcGuayOkay, see you folks. Hopefully that'll get the ball rolling...
18:55:03bluebrotherLlorean: I'll vote for that.
18:56:26Lloreanbluebrother: Though personally, I'd like some clever scheme to require they police themselves. For example "if any user top posts five times (with some leeway given for the occasional accident), we'll shut down the list, but staff will no longer provide rules reminders" so they can see for themselves the difficulty in keeping it up.
18:57:32*Llorean likes crazy schemes.
18:57:51bluebrotheryep, that could work. At least I think it's kinda the only way to get it work.
18:58:28bluebrotheras of now, I'm considering to write a last mail to that thread and afterwards unsubscribe. This whining is really a PITA.
18:59:18bluebrotherI'd also require a real name policy on the tracker ...
18:59:18 Join ender` [0] (
19:01:01Lloreanbluebrother: I wouldn't mind that at all.
19:01:10 Quit ompaul (Client Quit)
19:01:13LloreanNobody's presented me with a good reason why they can't have their real name associated with Rockbox.
19:01:36bluebrother"privacy reasons" :)
19:02:04bluebrotherbut in fact I consider not using your realname on a public list as impolite. Guess I'm not the only one ...
19:13:59 Quit hannesd (Read error: 113 (No route to host))
19:15:30 Nick soap is now known as Soap_OutOfTown (n=soap@rockbox/staff/soap)
19:16:31gevaertsthebishop: rockbox is not (and has never been) a modification of the original firmware
19:17:17thebishopgevaerts, right. but it sounds like the firmware of 6G would have to be modified for Rockbox to work, right?
19:17:32thebishopeither that or fool for encryption
19:17:40thebishop*the encryption
19:17:59gevaertsthebishop: it's a bit more complicated. You need to find out how the firmware is validated. That finding out is often done by modifying the OF
19:18:12 Join bertrik [0] (n=5a80bda4@gateway/web/cgi-irc/
19:18:30bluebrotherthebishop: no. We need to add our own bootloader. But if the hardware only runs code that is signed or encrypted we need to do that with out bootloader too
19:19:27 Join hannesd [0] (
19:19:34thebishopbluebrother, but writing your own bootloader doesn't get anywhere if the firmware won't run it
19:20:11bluebrotherdepending on what firmware you're referring to, yes. The boot mode needs to run our bootloader.
19:20:37 Quit desowin (Excess Flood)
19:20:44bluebrotherfor example, the h100 series simply executes code from flash. Thus we need to implant our own bootloader to the flash and make the original code jump to that
19:20:57 Join desowin [0] (n=desowin@atheme/member/desowin)
19:21:03gevaertsmcuelenaere: are you sure that "licensie" is correct? I would expect "licentie"
19:21:18bluebrotherthe older ipods load the bootloader from disc, so it's sufficiant to modify that.
19:21:30bluebrotheri.e. replace it with our own one.
19:21:52mcuelenaeregevaerts: woops, you're right :)
19:22:14mcuelenaeregevaerts: will you correct it or shall I?
19:22:20gevaertsmcuelenaere: I can do it
19:30:43 Join CaptainKewl [0] (
19:30:50 Join jhulst__ [0] (
19:31:17gevaertsbluebrother: "You need to restart the application for the changed language to take effect." Wouldn't 'the new language' be better?
19:32:13bluebrotherpossible, but then you need to rephrase it a bit more. A language doesn't take effect, does it? "to get applied" sounds much better to me.
19:32:38bluebrotherbut there are several string I'm not happy with. I just didn't want to start a rework before releasing.
19:33:54 Quit stoffel (Read error: 113 (No route to host))
19:42:37gevaertsmcuelenaere: I committed some updates to the dutch translation. Yell if you disagree with them :)
19:43:31mcuelenaeregevaerts: you've made a <translation type="unfinished">
19:43:48mcuelenaerethe punctuation probably didn't match with the English version
19:43:55*gevaerts tries to claim that this was intended
19:44:30mcuelenaereah no, I think you just pressed ENTER instead of CTRL+ENTER
19:45:46gevaertsOK. fixed. I'll commit again
19:46:47 Nick DarkFox is now known as HellDragon (n=jd@unaffiliated/helldragon)
19:47:22 Part J-23
19:49:02gevaertsmcuelenaere: I'm not entirely happy about the TTS and Talk term. Unfortunately I have no better ideas
19:50:24mcuelenaeregevaerts: yes I know, I tried to be a bit consistent in the naming of one of the 2 (I think it's Talk), but still it isn't good
19:51:12mcuelenaeregevaerts: which tool did you use to edit the file? Linguist?
19:51:58mcuelenaereif you go to CreateVoicFrm -> TTS profile: do you also see the question mark?
19:52:17mcuelenaerebecause that's what I meant with not matching with the original punctuation
19:52:38gevaertsah, yes.
19:53:22mcuelenaereI wonder though why Linguist doesn't report it..
19:53:38 Join J-23 [0] (
19:53:45gevaertsIndeed. It's a bit annoying to have to look everywhere
19:54:05mcuelenaerenormally, it marks it with a Question mark in the left panel
19:55:16mcuelenaeresomething what wonders me too: if you save the translation with Linguist it adds <defaultcodec></defaultcodec> but if you do "lupdate" it removes this line..
19:56:00gevaertsI fixed two punctuation differences. There's one remaining, but that one is wrong in the original
19:56:23 Join stoffel [0] (
19:56:33 Quit SirFunk_ (Read error: 104 (Connection reset by peer))
19:57:12mcuelenaeregevaerts: commit away ;)
19:57:33 Join SirFunk [0] (
19:58:23mcuelenaerejust one more commit and the Dutch translation of RBUtil rules the frontpage :)
19:58:42bluebrothermcuelenaere: seems to be really important ;-)
20:05:06gevaertsbluebrother: what's the policy on fixing typos in the source ? Do I commit updates for all languages to make the question mark go away, or just the ones I actually understand?
20:05:50bluebrotherif it's a change in the source string you can just change the ts xml and be fine.
20:06:03bluebrotherso there's no reason not to update the languages you don't speak
20:06:07 Quit jhulst__ (Read error: 60 (Operation timed out))
20:08:24 Join dataangel [0] (
20:08:32*gevaerts thinks that he found an easy way to increase his commit count
20:10:35 Part jon-kha ("[IRSSI] often imitated, never duplicated!")
20:10:45 Join jon-kha [0] (
20:14:02dataangelWhat do I have to do to set background color? I tried: rb->lcd_set_background(LCD_RGBPACK(0,0,0)); rb->lcd_clear_display(); rb->lcd_update(); /* But the rockbox logo still stays in the background */
20:14:48 Quit thebishop (Remote closed the connection)
20:15:31Lloreandataangel: The Rockbox logo?
20:16:27 Join jhulst__ [0] (
20:16:32dataangelLlorean: Yeah, the default background when you're using the menus and stuff has a vertical rockbox logo in the lower right corner
20:17:02dataangelactually I just tried red and setting the background color doesn't seem to have any effect at all, hrm...
20:17:04Lloreandataangel: Ah, background image. I do believe you have to unset the image before you can set a color, since the image takes precedence. I do not know how this is accomplished under the hood, though
20:21:42dataangelfound it :)
20:23:46amiconnThis behaviour might change soon
20:23:48 Quit davina (Remote closed the connection)
20:24:10Lloreanamiconn: Oh?
20:24:51amiconnMaking the usage of a backdrop (if it exists) part of the drawmode
20:25:20amiconnThis will open the possibility to decide on a per-viewport basis
20:25:38amiconnIt will also make it easier to further optimise mono bitmap drawing
20:25:51 Join davina [0] (
20:26:11LloreanI like "per-viewport basis" a _lot_
20:27:53amiconnI already have a partial implementation ready. lcd_puts_style_offset still needs work (and the greyscale drivers aren't done yet either)
20:28:11 Join Strife89 [0] (n=michael@
20:28:33dataangelDo ticks correspond to any unit of time?
20:28:56Strife89I am attempting to build my own version of Rockbox using the instructions on this Wiki page:
20:29:07 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
20:29:27Strife89However, I am stuck. I cannot execute the configure script.
20:29:40scorche|shwhy not?
20:29:59 Part pixelma
20:30:13Strife89I'm probably just not typing it in right, but I have no hints as to what.
20:30:15bluebrotherx bit not set?
20:30:32Lloreandataangel: I believe you'll need to use the HZ value to convert ticks to time.
20:30:35Strife89Let me pastebin my terminal.
20:30:47 Quit amiconn (" bbl")
20:31:09 Quit bertrik ("CGI:IRC")
20:32:15Strife89Here's my recent terminal output.
20:32:25Strife89I am a complete n00b, so.....
20:32:50bluebrotherStrife89: you're in the wrong directory
20:33:05bluebrotherafter line 9, you should've done: "../tools/configure"
20:33:37Strife89Alright, hang on and I'll try it.
20:33:53saratogacould someone test my WMA improvements on Coldfire?
20:33:57bluebrotherand you can't simply call "configure". If you're in the tools folder you'd need to call it as "./configure". But using it from that folder will produce only headaches, not the expected result
20:34:25Strife89michael@CarrDesktop:~/rockbox/build$ tools/configure
20:34:25Strife89bash: tools/configure: No such file or directory
20:34:47bluebrotheryou missed the leading dots ...
20:34:52saratogathats not how unix paths work
20:35:04bluebrotherplease read the _exact_ line I gave you.
20:35:10Strife89Sorry. :(
20:35:44Strife89Aha. :)
20:36:11Strife89And I just enter the number of the device?
20:36:29bluebrotherwell, the selection should be self-explanatory ... ;-)
20:36:29scorche|shas the page says...
20:36:45Strife89Okay, thanks.
20:37:27saratogaanyway if someone gets a chance to test that patch, I'd mostly like to know that it plays WMAs correctly without any glitching
20:38:23Strife89Might try it, when I get to applying patches.
20:38:57Strife89Drat, arm-elf-gcc is not in my path, so.....
20:39:22saratogaStrife89: if you're using arm-elf-gcc, you're probably not a Coldfire user
20:39:40saratogathough i guess i don't mind anyone finding bugs on arm either
20:39:50Strife89I'm building for a c250; Ubuntu 7.10 is my OS
20:40:12saratogathat would make you ARM
20:40:42Lloreansaratoga: Would these be obvious glitches (skipping, popping, etc) or subtle ones (requiring knowledge of what it's supposed to sound like)?
20:41:14LloreanNot that i can test, my H100's charger is missing.
20:41:37LloreanStrife89: You already told us that, you need to have the cross-compiler properly installed
20:42:09saratogaLLorean: probably obvious stuff like pops and blasts of static
20:42:21gevaertssounds dangerous...
20:42:23saratoga(keep your volume turned down low!)
20:42:26PaulJamsaratoga: i could test on my h300
20:42:36Strife89Llorean: Beg your pardon, but I have no idea what to do about that.
20:43:04saratogaStrife89: did you follow the wiki page explaining how to install the crosscompiler?
20:43:32Strife89Yeah, I've got that up.
20:43:40LloreanNo, you don't
20:43:41***Saving seen data "./dancer.seen"
20:43:51Strife89Tabs, my friend.
20:43:53bluebrotherwell, you obviously missed at least the last part about PATH
20:44:01LloreanStrife89: Oh, I thought you meant you've done it.
20:44:08saratogaanyway, if any problems turn up, I'll check the logs
20:44:19saratogaor maybe i should just open an FS entry for it
20:45:36Strife89So where do I execute $ which arm-elf-gcc ?
20:46:16scorche|shStrife89: why do you need to do that?
20:46:38Strife89Never mind.
20:47:07Strife89I'll try $ make for now, as is.
20:47:44 Join MethoS_ [0] (
20:48:22Strife89But of course it seems to have failed.
20:48:40LloreanStrife89: Are you at all familiar with general Linux terminal use?
20:48:44bluebrotherof course. If the compiler is not in the path running make will fail
20:49:07Strife89Tch. Barely. I'm trying a crash course here.
20:49:16bluebrotherStrife89: may I suggest you dropping the Rockbox stuff for now and reading up on basic linux usage first?
20:49:30LloreanStrife89: Then stop "crashing" and go back to "coursing", specifically find a general tutorial on terminal use, and make sure you know what you're doing *before* you do it.
20:49:37LloreanThat way you're less likely to get stuck constantly
20:51:31 Join Thundercloud_ [0] (
20:51:55 Quit saratoga ("CGI:IRC")
20:52:18PaulJamsaratoga: i get a warning when compiling for h300 with the patch.
20:57:49 Join atrus [0] (
21:02:16 Quit Rob2222 (Read error: 104 (Connection reset by peer))
21:04:04 Quit Thundercloud (Read error: 110 (Connection timed out))
21:06:17J-23hmm, will it be offtopic here if I ask where and what files should I copy to my player to install Chex Quest Doom mod on my Rockbox player?
21:06:19 Quit MethoS- (Read error: 110 (Connection timed out))
21:06:36 Quit ch4os ("leaving")
21:07:03gevaertsJ-23: That's sneaky ;)
21:07:27 Join webguest72 [0] (n=40deb184@gateway/web/cgi-irc/
21:08:25webguest72when i turn my 20gb rocbox on it tells me "operating system is not found" what do i do to fix this can anyone tell me?
21:09:05Lloreanwebguest72: It sounds like a message from the original firmware.
21:09:05 Quit stripwax (Read error: 104 (Connection reset by peer))
21:09:08gevaertswebguest72: "20gb rocbox" ?
21:09:18 Join stripwax [0] (
21:09:23stripwaxAnyone know how to get more functions listed in the map file? Looks like, to my untrained eye, the map file you get on arm lists only the functions that have some external linkage in a given file, and file static functions are not listed (or don't seem to be.. I'm looking at and the only function listed from layer3.o is mad_layer_III, which is some kind of entry point there)
21:09:38Lloreanstripwax: You were asking if I was around earlier?
21:10:15Lloreanwebguest72: We don't know what player you have, so it's pretty much impossible to help more. It sounds like you may not have a complete installation.
21:10:15 Quit webguest72 (Client Quit)
21:10:18scorche|shstripwax: do you have your shirt?
21:10:20stripwaxLlorean - was not important - just a question on channel guidelines, and what acronyms are acceptable in common usage
21:10:34stripwaxscorche - I do, but I'm not wearing it today :)
21:10:36Lloreanstripwax: Generally speaking, "only the ones in the project glossary page" :)
21:10:42stripwaxLeft my sticker behind though :(
21:10:56stripwaxLlorean - heh.
21:11:15stripwaxscorche - did you end up with a spare?
21:11:28scorche|shnot a spare...a previously worn one
21:11:35*Llorean thinks the ProjectGlossary should be linked in the topic too, with an "Acronyms Here: Link" bit
21:12:06stripwaxscorche - odd. i kept mine on, and wore it home, so I know it's not mine
21:12:16*mcuelenaere thinks he isolated the LCD init routine
21:12:37 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
21:12:52stripwaxsaratoga - (for the logs), looks like a profile build works fine on arm, other than the map file having nothing like enough detail to do anything useful with the profile output :)
21:12:53*shotofadds is missing a t-shirt
21:13:05shotofadds...and a sticker
21:13:09*J-23 copied doom.wad file from CQ's main directory to /.rockbox/doom/rockdoom.wad, but Doom plugin returns "Missing base WAD" :/
21:13:24scorche|shshotofadds: thats what we to GodEater_ when he gets done watching TV for how to get it ;)
21:13:56gevaertsJ-23: Did you read ?
21:14:15J-23gevaerts: yes, but I didn't understand it.
21:14:53 Join spiorf [0] (
21:15:07gevaertsJ-23: basically you need the rockdoom.wad that's linked from there, and then any .wad files from variants you want to use
21:16:30shotofaddsscorche|sh: it seems Lambda's m200 didn't like being pulled apart in the pub .. it's now missing about 8 LCD lines :/
21:17:23scorche|shshotofadds: eek...i never checked it since it was shipped, so it could have been a number of things...did you check the connector?
21:18:16 Part Strife89
21:18:43LloreanJ-23: Rockdoom.wad should never, EVER be replaced.
21:20:57J-23hmm, I downloaded rockdoom.wad to /.rockbox/doom/ and moved CQ's WAD to /.rockbox/doom/addons, but RockDoom still returns "Missing base WAD!".
21:21:15J-23but I see my player often losses files :/
21:21:40bluebrotherJ-23: you're repeating yourself ... in that case, tried chkdsk-ing the filesystem?
21:22:05LloreanJ-23: you need to get doom working normally, first.
21:22:05*bluebrother still wonders why his m200 doesn't seem to go to usb boot mode :/
21:23:09shotofaddsscorche|sh: applying some pressure to the top of the lcd makes some of the 'dead' lines re-appear, so I guess it could be the connector/ribbon. but I can't get at it close enough to try and tweak it
21:23:28PaulJamisn't the doom.wad or doom2.wad in the .rockbox/doom/ dir also reqired?
21:23:30Lloreanbluebrother: Another rbutil feature request: Since speech files are player-dependent anyway, can we take away the "speex" and "lame" options, and just make them depend on what player they're making the file for?
21:23:35LloreanPaulJam: Yes.
21:23:43LloreanAn IWAD is required.
21:23:53LloreanThough I wouldn't be at all surprised if the Chexquest wad can be renamed doom.wad
21:23:54J-23I tried fscking Rockbox partition, but fsck returns "Invalid disk format in boot sector."
21:23:56LloreanOr perhaps doom2.wad
21:24:06LloreanSince it is technically an IWAD.
21:24:08shotofaddsbluebrother: is that with the daughterboard removed, or by shorting the USB pin? the latter is sometimes temperamental on mine, but removeing the board always works
21:24:15stripwaxJ-23 - you mean fsck.vfat , right?
21:24:35J-23no, dosfsck :/
21:24:54gevaertssame thing
21:25:01bluebrotherLlorean: that is already implemented
21:25:08J-23hmm, when I runned it as fsck.vfat, it returned "Logical sector size is zero."
21:25:32bluebrothershotofadds: I've removed the daughterboard but nothing happens.
21:25:38gevaertsAre you sure you're running it on the rockbox partition?
21:26:15shotofaddsah. nothing will show on the screen, but it'll be picked up on the PC as an unrecognized usb device
21:26:16Lloreanbluebrother: How do people keep making bad voice files then?
21:26:27Lloreanbluebrother: Or is this in just the newest version?
21:26:31bluebrotherLlorean: it's in the upcoming release ;-)
21:26:45 Quit petur ("switching")
21:26:49 Join petur [50] (n=petur@rockbox/developer/petur)
21:27:00bluebrotherwhich I intend to release today. I just found a small issue with that and decided to add a tiny workaround first.
21:27:39 Join Zagor [242] (
21:29:01Zagorwouldn't it be helpful to have a simplified changelog on CustomWPS? to list when tags are added, removed or changed.
21:29:38J-23hmm, I think it's Rockbox partition - I see .rockbox dir on it.
21:30:06J-23so what should I do when it's damaged?
21:30:18gevaertsIs it still mounted ? If so, unmount it first
21:30:23scorche|shshotofadds: cant get close enough?
21:31:04Lloreanbluebrother: Okay, just making sure I wasn't being stupid and requesting something that's already in a release version, that I completely overlooked. :)
21:31:20LloreanZagor: Is that possible?
21:31:30ZagorLlorean: I mean written manually
21:31:33 Quit spiorf (Read error: 110 (Connection timed out))
21:31:35bluebrotherLlorean: I noticed that forum thread and wondered myself if 1.0.5 already included that feature ;-)
21:31:41LloreanZagor: I could see it as being useful.
21:32:01Lloreanbluebrother: I knew 1.0.5 didn't. I just wasn't sure if 1.0.6 had made it out when I wasn't looking, and I'd overlooked it. :)
21:32:04bluebrotherjust force devs to edit a changelog when they change it. Works quite fine for rbutil ...
21:32:09 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
21:32:29saratogastripwax: i don't think map files are supposed to show static functions
21:33:03saratogathe perl script gets the non-static functions by objdump on the actual codec binaries
21:33:13saratogabut i never had any luck finding most of them even doing that
21:33:27saratogaor at least the addresses didn't make sense to me when i tried
21:34:27shotofaddsscorche|sh: I mean I'd need to take the LCD off the board to have a proper look, but I don't know if that's even possible.
21:34:43stripwaxsaratoga - yep, was just looking into objdump in fact. but the addresses look good to me. i'm not using the perl script though (I seem to remember it only works for the coldfire outt
21:34:44shotofaddsCurrently on the "USB Connected" screen most of the "USB" is missing :/
21:34:53scorche|shshotofadds: i cant imagine it not being able to, but i suppose it is possible
21:35:07LloreanZagor: Honestly, if we had a "WPS Design" manual section, we could integrate that file so that we had a simplified changelog, and an easier official place for people to look up the tags?
21:35:22bluebrothershotofadds: the strange thing is that my PC doesn't pick the device up −− lsusb doesn't show anything, and windows doesn't tell about an unknow usb device.
21:35:24PaulJamsaratoga: did you read what i wrote shortly after you left?
21:35:35bluebrotherany hints? Do I need to flip some switch first or something like that?
21:36:12shotofaddsscorche|sh: yeah, I'll try lifting it off sometime. but for now I can just press down in the right place if I need to see those lines...
21:36:17J-23it's unmounted and fsck.vfat returns "Logical block blah blah blah"
21:36:18bluebrotherconnecting the player in "working" mode itself works fine
21:36:21J-23good night!
21:36:22 Part J-23
21:36:30saratogaPaulJam: yes I fixed that warning a while ago and didn't bother to update the patch since its not important
21:37:45saratogayou can fix it yourself just by casting the buffers on that line to (int*) instead of char* or whatever they default to (Tremor and libwmai disagree on how they declare their buffers)
21:38:18PaulJamsaratoga: the one WMA song that i have sounds fine with the patch.
21:38:28shotofaddsbluebrother: if you've taken the daughterboard off it should go straight into USB boot mode and show up in lsusb as 140e:b021. there's no extra switch or anything.
21:39:58shotofaddswhich hardware revision do you have?
21:40:06saratogaPaulJam: thanks a lot
21:40:21saratogacan you run test_codec on your file with and without the patch too?
21:40:33*bluebrother has a segfault of rbutil :(
21:41:02 Join hannesd_ [0] (
21:43:07 Join Thundercloud [0] (
21:43:13stripwaxsaratoga - in my profiling, mpa.codec spends 882k ticks in synth_full, 202k ticks in mad_bit_skip, 187k ticks in III_imdct_s, 503k ticks in III_decode, everything else <100k ticks
21:43:59 Quit perrikwp (" ajax IRC Client")
21:44:58saratogastripwax: thats about in inline with what I found
21:45:04 Quit hannesd (Read error: 145 (Connection timed out))
21:45:05 Nick hannesd_ is now known as hannesd (
21:45:19stripwaxThe addresses make sense
21:45:23saratogawhat bitrate file is this ?
21:45:42 Quit bluebrother (Nick collision from services.)
21:45:44stripwaxabout three files, average of something like 160kbps. Not rigorous I know.
21:45:45 Join bluebroth3r [0] (n=dom@rockbox/staff/bluebrother)
21:47:03saratogathis is compared to objdump and the map file addresses?
21:47:05 Nick bluebroth3r is now known as bluebrother (n=dom@rockbox/staff/bluebrother)
21:47:12stripwaxobjdump addresses, yes
21:47:35stripwaxwhich ties out with the map file, for the non-static functions that do show up in it
21:47:36saratogaok i'll have to take a look at the perl script and see why its not working then
21:48:22saratogaif you get a chance, looking at that 96kbps file on the test_codec wiki page might be interesting too
21:48:23stripwaxBtw there is a bug in profile.c , depth++ in print_recursive should be ++depth, otherwise it will just print out 0000 as the depth in all lines
21:51:37saratogaare you going to commit a fix for it?
21:51:53stripwaxI don't have commit access, so , no :)
21:52:31bluebrotherstripwax: you can "commit" it to the tracker ;-)
21:52:44stripwaxfor a four-character edit? :)
21:53:06 Nick midkay_ is now known as midkay (
21:53:11saratogaah didn't realize you didn't have SVN
21:53:21stripwaxyep, still don't.
21:53:21PaulJamsaratoga: without the patch: 249,25% and with the patch: 298,75%
21:53:22bluebrotherbetter than feature requests like "port to Nano 3G" ;-)
21:53:39saratogabut yeah, might as well create a patch so that someone sees it (i'm kind of hesistant to blind commit that and I don't have enough time to dig into it today)
21:54:07saratogaPaulJam: wow thats really good
21:54:16 Quit stoffel ("leaving")
21:54:22stripwaxok, no worries. there's a handful of other things that need to be changed to get profile builds actually working on arm anyway, so will create a patch with everything in it
21:54:37saratogathats an 8 MHz speedup on Coldfire
21:55:12saratogastripwax: sounds good
21:56:01stripwaxlooks like the perl script either doesn't like the format of the arm mapfile, or assumes it lists all address (which it won't)
21:56:09saratogaPaulJam: what birate?
21:57:19PaulJamsaratoga: 128kbps (it is one of the windows example music songs)
21:59:28 Quit Thundercloud_ (Read error: 110 (Connection timed out))
21:59:47 Quit sarixe ("(EE) Failed to load "quit" module")
22:00:11saratogag_128k is the MS example for WMA9, but I guess its not what you're using
22:00:52PaulJamno, it was "david byrne - like humans do"
22:04:43saratogajust playing with this patch, and it almost but not quite decodes WMA unboosted
22:04:50saratoga(on the Sansa)
22:05:01saratogaerr for 128k files
22:05:11LloreanStill, a good target.
22:06:42 Quit merbanan (Remote closed the connection)
22:07:45saratogathe numbers on the audio buffering debug screen seem a bit odd
22:08:01saratogatest codec reports ~32 MHz for this file, but that debug screen claims 41MHz
22:08:28saratogawith no DSP, that would mean the overhead from rockbox and drawing that screen was 9 MHZ
22:08:34saratogawhich seems a bit too high
22:09:42LloreanDid you go in after buffering was finished, etc?
22:09:53LloreanAnd how much time is spent dealing with the PCM buffer, etc?
22:10:04 Join Rob2222 [0] (
22:10:48saratogaLlorean: I'm not sure if it was buffering, i just assumed that on a sansa that would be quick, though maybe not
22:11:52 Join Buschel [0] (
22:12:21PaulJamsaratoga: on h300 it is similar, in buffering screen i get ~54 MHz (entered after buffering finished) and in test_codec ~41,6MHz.
22:12:26Buschelsaratoga: you're making good progress on WMA speed :)
22:14:41saratogaBuschel: I remember you looked into rockbox overhead a little
22:14:49saratogahow much does rockbox itself use?
22:14:55saratoga(above what the codec needs)
22:14:56bluebrotherZagor, Bagder: can you move and .zip to the download server?
22:16:12bluebrotherlet's see how many bugs are left in that release ;-)
22:16:23Buschelregarding the buffering screen -> dsp needs some MHz (+2.5MHz on PP for mpc, i do not know whether there is some resorting needed for WMA). the screen update will also take some MHz
22:17:04Zagorbluebrother: done
22:17:57bluebrotherthanks. Updating the wiki ...
22:18:00saratogai think overhead from DSP should be low for WMA since the output buffer is in IRAM and i believe its in the desired format already (interleaved 32 bit PCM if I recall correcly)
22:18:08Buschelsaratoga: you can for the influence of the screen update via doing 2 measurements (1. @10fps = default, 2. @0.1fps)
22:18:48Buschelsaratoga: then for WMA the overhead should be the same than waht I've measured = ~2.5MHz
22:19:20gevaertsZagor: shouldn't the date for "DevCon 2008 is over." not be 2008-06-29 instead of 27 on the front page ?
22:19:51Zagorgevaerts: definitely. fixed.
22:20:16gevaertsNot the most important of problems of course :)
22:20:59saratogaBuschel: if we use 5-10MHz on GUI, then no wonder our battery life isn't as good as the retail firmware
22:23:07Buschelsaratoga: 10fps is used for debug screen −− and it's needed for the bar graphs. 5fps is used by the WPS −− always, independent of changed/unchanged content
22:23:59Buschelsaratoga: that's why battery benchs with "no WPS" make sense −− you totally switch off any gui updates
22:24:51 Quit Thundercloud (Read error: 110 (Connection timed out))
22:25:54saratogaBuschel: ah I see, so the debug screen has 2x the overhead
22:25:57saratogasounds reasonable
22:26:41 Quit saratoga ("CGI:IRC (EOF)")
22:26:58stripwaxsaratoga - profiling lame_96 now. oh, you'll read about it later :)
22:27:28 Join Lear [0] (i=chatzill@rockbox/developer/lear)
22:28:08mcuelenaereZagor: I think you still need to copy to
22:29:08Zagormcuelenaere: it's there, it just hasn't been grabbed by all mirrors yet. use to view the main server.
22:29:46mcuelenaereah ok
22:30:02 Quit dataangel (Read error: 110 (Connection timed out))
22:30:07 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
22:30:25*mcuelenaere didn't know Rockbox used such a sophisticated mirroring system ;)
22:30:48 Quit petur ("discworld reading, then Zzzz")
22:40:10stripwaxlame_096: 525k ticks in synth_full, 104k ticks in mad_bit_skip, 466k ticks in III_decode, and no appreciable time in III_imdct_s
22:41:45 Join Thundercloud [0] (
22:42:29 Join LardMaster [0] (
22:42:37stripwaxlame_192: 580k ticks in synth_full, 128k ticks in mad_bit_skip, 477k ticks in III_decode, also no appreciable time in III_imdct_s. curious
22:43:15 Quit LardMaster (Client Quit)
22:43:43***Saving seen data "./dancer.seen"
22:44:40 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
22:45:44 Join lard-Vader [0] (
22:46:43lard-Vadercan I have wiki write permissions to note that all you need to do is install libsdlX-dev and get the svn to compile it, etc if you are on debian?
22:47:05Llorean"compile it"?
22:47:11LloreanDo you mean the simulator?
22:47:18LloreanWhat page are you referring to?
22:47:27LloreanYou don't know?
22:47:30lard-VaderI'm a little new at this
22:47:32 Quit K4rP4D (Read error: 104 (Connection reset by peer))
22:47:38lard-Vaderbut it seems like I built most of it
22:47:42LloreanI really would rather you not update the wiki with information if you don't even know what you're going to be updating...
22:47:58lard-VaderI just did
22:48:02stripwaxAre the timer ticks in the profiling builds a fixed time interval or does the duration change as the clock scales?
22:48:10lard-Vaderit tells me "you need libsdl installed"
22:48:19lard-VaderI installed it.
22:48:34LloreanIf you're trying to build Rockbox itself, you don't need SDL.
22:48:47lard-Vaderthat explains it.
22:48:49LloreanWhat on Earth are you doing, and what instructions are you following?
22:48:56 Join Thundercloud_ [0] (
22:49:21lard-VaderPlatform set to h300
22:49:21lard-VaderBuild (N)ormal, (A)dvanced, (S)imulator, (B)ootloader, (M)anual: (Defaults to N)
22:49:26lard-VaderI chose simulator
22:50:41lard-Vadersorry... I semi know what I am doing
22:50:54LloreanProbably not.
22:50:55scorche|shyes...why choose that if you wanted a normal build?
22:51:02LloreanSince you shouldn't be running configure from the directory it's located in.
22:51:11LloreanHave you read the actual instructions for building?
22:51:39lard-VaderI thought I did.
22:51:56LloreanI asked you what instructions you're following a few times now.
22:52:38lard-Vaderbut I didn't follow the previous instructions
22:53:37LloreanJust the "Compiling" section makes it clear you need to create a build folder, and use ../tools/configure rather than running ./configure from within the tools folder.
22:53:49lard-VaderI did do that
22:53:54lard-VaderSorry if I was unclear
22:53:59 Join culture [0] (
22:54:12LloreanWell, saying you did "./configure" could be considered rather misleading if that's not what you did...
22:54:49lard-Vadersorry... I should have said I ran configure
22:54:58LloreanWell then I would've asked you from where. :)
22:55:16lard-VaderWEEEEEEEEP. Don't run this configure script within the tools directory.
22:55:16lard-VaderIt will only cause you pain and grief. Instead do this:
22:55:26lard-VaderAlso, it prints out this nice message if you do
22:55:27LloreanYou cannot build rockbox without the necessary cross-compilers. You cannot build the sim without the appropriate SDL libraries. But they aren't necessary for a normal build.
22:55:50scorche|shand please stop just pasting things in here
22:55:51bluebrotherimportant thing when programming: be exact. Especially when looking for help
22:55:53 Join vort3x [0] (n=vortex@unaffiliated/dfa001)
22:56:00LloreanIt may print out that message, but people have been known to work around it.
22:56:23 Join Schmogel [0] (
22:56:55*Bagder likes the WEEEP error message! ;-)
22:57:55lard-VaderWhoever tries to work around such a wonderful message has a demented mind.
22:58:12Lloreanlard-Vader: People are strange.
22:58:24BagderI think it basically happens because configure writes Makefile in the current dir
22:58:39Bagderso if you run configure in a funny dir, you get the original Makefile overwrittten
22:58:54Bagderthus no WEEEP anymore
22:59:08Bagderjust failure ;-)
23:00:11 Quit Thundercloud_ (Remote closed the connection)
23:00:43 Join Thundercloud_ [0] (
23:00:48*Llorean thinks if ./tools/configure is run, not ../, it should create subfolders for every possible build target (bootloaders, manuals, sims, normal, at least) and spend a week building once someone runs make, explicitly for those who skip the instructions on their first try.
23:01:02Llorean*run from the Rockbox source folder, obviously
23:01:10 Quit Buschel ()
23:01:14*gevaerts would like that for really complete test builds
23:01:34Lloreangevaerts: That's just what "commit" followed by "oops" is for.
23:02:20amiconnBagder: Speaking of that - would it be possible to extend the distributed build system to do manual builds if changes to the manual are committed?
23:02:38Bagderof course, but it would probably take some poking
23:02:51Bagdersince I bet very few of the servers have that setup
23:02:59 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
23:02:59amiconnOf course this will probably need a separate list of machines, as it has different requirements that target builds or sims
23:03:25bluebrotherdomonoky: you can start building a mac binary ;-)
23:03:47bluebrotherBagder: having a way to access the manual build logs would be nice too.
23:04:11domonokybluebrother: not today, will do it tomorrow...( i just arrived at home)
23:04:13amiconnPerhaps it would be good to do a similar thing for voices once we start distributing multiple languages
23:04:17pixelmathough they can be somewhat long...
23:04:56bluebrotherno hurries −− the last release was over 2 month late ;-)
23:05:39 Quit Thundercloud (Read error: 110 (Connection timed out))
23:06:22Lloreanamiconn: If I understand correctly, the plan is not only multiple languages but multiple engines.
23:06:29 Join MethoS- [0] (
23:06:32LloreanThat's a lot of building
23:06:39amiconnyep, exactly
23:08:43 Quit Arathis ("Bye, bye")
23:09:12 Quit desowin ("KVIrc 4.0.0 Insomnia")
23:19:13 Quit Thundercloud_ (Remote closed the connection)
23:19:16 Quit Rondom ("Ex-Chat")
23:19:49*preglow wonders how apples api for voice works
23:20:03preglowthe voice that came with leopard isn't half bad
23:20:31Zagorpreglow: but are you allowed to distribute voice files created with it?
23:20:50preglowZagor: a fair question
23:21:04 Join MethoS-- [0] (
23:22:02preglowi haven't even bloody managed to make a working cross-compiler yet under os x, so i guess i should stay away from voice until i have a working dev environment anyway :)
23:22:46 Quit domonoky (Read error: 104 (Connection reset by peer))
23:24:18 Quit MethoS_ (Read error: 110 (Connection timed out))
23:31:33 Quit HauDyr ("VI ses")
23:34:34MarcGuaySorry about the extra diffs in my last commit. I forgot I'd played around a bit with the c100 stuff - no major changes.
23:34:49stripwaxat fullspeed (i.e. test_codec), 96kbps lame spends about 300k ticks in each of synth_full and III_decode, 65k ticks in mad_bit_skip; 128kbps spends 426k ticks in synth_full vs 334k ticks in III_decode, 77k ticks in mad_bit_skip; 192kbps spends 426k ticks in synth_full vs 351k ticks in III_decode, 94k ticks in mad_bit_skip
23:35:19stripwaxvery interesting. so synth_full on 128kbps example file is no faster than 192kbps example file
23:35:28preglowoh no
23:35:34preglowthe synth step is the same anyway
23:35:39preglowthe amount of data in and out doesnt change
23:35:49preglowthe only part that cares about bitrate is entropy coding
23:36:19stripwaxWhy is 96kbps much faster?
23:37:16pixelmaMarcGuay: you could specify the files to commit or what I sometimes do - change into the manual (as an example) folder and commit from there
23:37:53stripwaxpretty much same performance (=200% realtime) for 128kbps thru 320kbps, and only 96kbps is noticeably faster
23:37:58MarcGuaypixelma: Aha, thanks for the tips.
23:38:09Learstripwax: lower sample rate at 96 kbps.
23:38:21LearThus, less data in/out.
23:38:51preglowmpeg audio has a configurable upper band number
23:38:54preglowi guess that is why bit rate matters
23:38:59stripwaxoh! well that explains it. but doesn't explain why coldfire *does* scale more noticabley
23:39:03preglowfor synth_full
23:39:39stripwaxalthough 400->444% realtime vs 577% realtime @ 96kps on H140, so same thing there too really
23:39:58 Quit jhulst__ (Connection timed out)
23:40:02Learsynth_full buffers not in iram?
23:40:07stripwaxI guess profiling this on arm hasn't uncovered anything we didn't already know, synth_full is slow on arm.
23:40:12stripwaxLear - not sure
23:40:13preglowcan't remember
23:40:25stripwax5022/4 have 32KB more iram than they did a couple months ago ;)
23:40:33preglowsynth_full is indeed slow on arm7 arm
23:40:38 Quit MethoS- (Read error: 110 (Connection timed out))
23:40:39preglowtons of 64 bit arithmetic
23:43:20preglowi remember i had a small look through ipod retailos for 64 bit arithmetic and couldn't find much
23:46:16preglowafaik, the musepack subband synth routine is quite fast now, would be fun if we could adapt that to libmad
23:46:30preglowthe functionality itself should be equivalent
23:49:11 Quit roolku ()
23:54:15mcuelenaeredoes anyone know in what format 16bit RGB data is?
23:55:04 Quit dionoea (Remote closed the connection)
23:55:11 Join dionoea [0] (
23:55:50mcuelenaerenever mind
23:57:18 Quit vort3x (Read error: 104 (Connection reset by peer))
23:57:41 Join vort3x [0] (n=vortex@unaffiliated/dfa001)

Previous day | Next day