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 2008-10-22

00:00:43*LambdaCalculus37 found one reference to the PID of the iPhone
00:00:50LambdaCalculus37The PID is 0x1222.
00:00:58LambdaCalculus37But that doesn't help me with the nano.
00:01:52mcuelenaereit's weird, but there 'hash' would be an image_attr_t (see meizu_dfu.c), but it's just half of that struct
00:02:50 Quit sarixe (Read error: 110 (Connection timed out))
00:04:01 Join krazykit [0] (
00:04:04 Join AJCantos [0] (
00:04:15mcuelenaerethey seem to use
00:04:49mcuelenaeregevaerts: does that make any sense to you?
00:05:01linuxstbmcuelenaere: Do you know how DFU mode works in general? Does it upload an image which is then installed by software already running on the device, or does it upload code which is run directly?
00:05:22 Quit matsl (Remote closed the connection)
00:05:36mcuelenaereit is similar to the TCC/Onda/.. USB mode
00:05:45mcuelenaere(I think)
00:05:50mcuelenaereso it should run code from RAM
00:06:09linuxstbOK, so it's basically a standard form of usb boot/
00:06:13gevaertsmcuelenaere: no idea 'm afraid
00:06:24mcuelenaerewhat I'm sure is that the code running the DFU mode is written in hardware
00:06:32mcuelenaereie it can't be changed with a fw upgrade
00:06:38gevaertslinuxstb: DFU doesn't actually specify very much about that if I understand it correctly
00:06:41mcuelenaerebut it can (and will) be Apple-customized
00:07:12 Join Riyonuk [0] (
00:07:16mcuelenaerelinuxstb: I think it is meant to upgrade firmware's on devices, but isn't used like that here
00:07:30linuxstbWhere is "here"?
00:07:31RiyonukHow do I debug .wps files, using the simulator, if I cant even play a song without it closing
00:07:54linuxstbRiyonuk: Did you copy the .rockbox folder from your ipod to the simulator?
00:08:04RiyonukNope, draged one song over
00:08:22linuxstbSo do you see any messages when you try and play a song?
00:08:38mcuelenaerelinuxstb: the whole iPhone/iPod touch unlocking/jail breaking stuff
00:08:53 Quit Strife89 ("Leaving")
00:08:53mcuelenaerethey misuse it to upload a RAM disk to the device
00:09:22Riyonuknope, the app just closes
00:09:48 Part Palintheus ("leaving")
00:10:16linuxstbRiyonuk: That's not normal... What kind of file is it? What OS are you running?
00:11:35Riyonukmp3, and windows xp
00:12:36 Join Zarggg [0] (
00:12:45Riyonukit just closes, LOL
00:13:57RiyonukI put the mp3 in a folder thats on the same directory as .rockbox
00:15:49Riyonukany idea >_>
00:15:59 Join FOAD_ [0] (
00:16:02LambdaCalculus37Put them in a different folder.
00:16:23LambdaCalculus37You mean to say that you placed your MP3s in the .rockbox folder?
00:17:09RiyonukI put the mp3 at
00:18:12 Quit n1s ()
00:18:33linuxstbRiyonuk: That should work fine. Does the simulator work OK apart from music playback? e.g. can you try some plugins?
00:21:05Riyonuklet me check
00:21:41Riyonukyeah, tried the matrix plugin, works fine
00:22:07 Quit bughunter2 ("bye")
00:23:20RiyonukI just want a decent WPS screen that works with my background, is that so much to ask :(
00:23:28linuxstbRiyonuk: Then I don't know. Where did you get the sim from?
00:23:45linuxstbRiyonuk: The sim works for everyone else - you're just unlucky I guess.
00:24:25RiyonukI hate my life XD
00:24:46RiyonukAre you good with making wps's?
00:25:05 Join fdinel [0] (
00:25:13 Join mib_eyfsrl [0] (i=4596a301@gateway/web/ajax/
00:25:25 Nick mib_eyfsrl is now known as kushal_12_27_200 (i=4596a301@gateway/web/ajax/
00:25:57 Quit meven (Remote closed the connection)
00:27:38linuxstbRiyonuk: No, I don't really care about WPSs and just use the default.
00:27:49 Quit FOAD (Read error: 110 (Connection timed out))
00:27:50 Nick FOAD_ is now known as FOAD (
00:28:25 Quit LambdaCalculus37 (" ajax IRC Client")
00:28:30RiyonukI would use the default, but cant read it, due to the background.
00:29:00 Quit kushal_12_27_200 (Client Quit)
00:30:07 Join kushal_1227200 [0] (i=4596a301@gateway/web/ajax/
00:31:18RiyonukIt works now, yet it says in the top left "Divide by zero at 0002DEC0 (0)"
00:31:39 Quit mcuelenaere ("Zzzzzzzzzzz")
00:31:41 Nick kushal_1227200 is now known as kushal_12_27_200 (i=4596a301@gateway/web/ajax/
00:31:58 Quit ender` (" On the other hand, you have different fingers.")
00:32:22 Join einhirn [0] (
00:33:44XavierGrfml: sorry but when you come in I am absent.
00:34:46 Quit tvelocity (Read error: 110 (Connection timed out))
00:36:03 Join tvelocity [0] (n=tony@
00:37:54Riyonukwhats divide by zero mean?
00:38:09 Quit AJCantos (Read error: 110 (Connection timed out))
00:41:36kushal_12_27_200I have a problem with my sansa c250. When I connect it to a computer, the computer no longer recognizes it. What should I do?
00:43:21pixelmaRiyonuk: did you try with another mp3? Does that very file play correctly on your Ipod?
00:44:01 Quit shotofadds ("Leaving")
00:44:14 Join AJCantos [0] (
00:45:05RiyonukYes and Yes
00:45:12RiyonukAnd I keep getting divide by zero
00:45:15RiyonukThis is retarded
00:45:51linuxstbRiyonuk: Where did you get the sim from?
00:46:06 Join lasser [0] (
00:49:20pixelmafor what it's worth, I just tried a self-compiled sim (recent SVN) under XP and it plays music just fine
00:50:25pixelmahmm... wait, didn't update to the latest... try again
00:52:59AJCantosHi guys, sorry to bother you, I'd like to help with the Rockbox player
00:53:25AJCantosI'd try to incude my name in the wiki, but I don't have rights to write
00:53:38AJCantosAnyone can help?
00:55:00pixelmadid you already register, if so what's your wiki name?
00:57:47pixelmaok, you should be able to edit pages now
00:58:00pixelmahope you don't intend to spam ;)
00:58:24AJCantosdon't worry, thanxxx!
00:59:07pixelmayou're welcome :)
01:01:46 Quit tvelocity (Read error: 110 (Connection timed out))
01:03:07***Saving seen data "./dancer.seen"
01:03:27RiyonukHmm, since Im getting /0 errors, should I get stable rock? Or bleeding edge?
01:07:44 Quit nuonguy ("Leaving")
01:08:06 Join nuonguy [0] (
01:08:37 Quit Nico_P (Remote closed the connection)
01:10:27pixelmamp3 playback indeed is broken in a current SVN windows sim. It doesn't crash for me, "just" tries to load the song, splashes "Codec failure" briefly and skips to the next
01:11:00kushal_12_27_200I have a problem with my sansa c250. When I connect it to a computer, the computer no longer recognizes it. What should I do?
01:12:06krazykitkushal_12_27_200, does it not boot into the OF?
01:12:29 Quit faemir (Remote closed the connection)
01:12:34pixelmasame with other formats I tried (adx, mod (exotic formats I know) and wma)
01:13:16Riyonukpixelma, can you see if the zelda theme works for you? or anyone for that matter?
01:13:35scorche|shit probably doesnt
01:13:40scorche|shmost of the themes there are broken
01:14:15linuxstbpixelma: Hmm, works fine for me on Linux.
01:14:36RiyonukBut why? I don't undestand
01:14:47RiyonukI checked each syntax by hand, all are there
01:15:43scorche|shRiyonuk: run it in the simulator with the −−debug-wps argument
01:15:57 Quit miepchen^schlaf (Remote closed the connection)
01:16:37 Join miepchen^schlaf [0] (
01:16:56kushal_12_27_200it boots in the original firmware, yes. krazykit
01:17:03 Join pabs_ [0] (
01:17:15Riyonukthis? debugwps.bat
01:17:17kushal_12_27_200I tried rebooting the sansa while it was connected to the computer but to no avail
01:17:21 Quit pabs (Nick collision from services.)
01:17:31 Nick pabs_ is now known as pabs (
01:17:40 Nick fxb is now known as fxb__ (
01:18:38krazykitkushal_12_27_200, even if you first boot into the OF then connect to the computer? and what OS are you running?
01:19:12pixelmalinuxstb: unfortunately I don't get useful debug output. I could only try to track it down by binchopping (that will take a while though...)
01:19:43linuxstbpixelma: Have you tried many different mp3s?
01:19:54pixelmaonce in a while I get a "bufopen: failed to add handle" when it tries to load different tracks
01:20:35kushal_12_27_200Intel macbook 10.4.11 primarily
01:20:36pixelmalinuxstb: also different formats as I already said
01:20:44linuxstbThe recent suspicious commits would be n1s's commit to remove the malloc buffer, and Zagor's commit to reorganise the metadata. I can't think why it would be Windows specific though...
01:21:08linuxstbpixelma: Which sim did you build? I build an iPod Color
01:21:10RiyonukUnexpected conditional char after token 2, hmm
01:21:21kushal_12_27_200I am on a win xp sp2 machine right now
01:21:24pixelmalinuxstb: c200, the metadata is displayed correctly
01:21:25RiyonukI'm guessing it's something to do with this "%P|pb.bmp|"
01:22:54linuxstbRiyonuk: Yes, the %P tag doesn't exist any more.
01:23:13krazykitkushal_12_27_200, perhaps try a different port on the computer, or a different computer altogether. i'm really not familiar with windows
01:23:21pixelmaRiyonuk: the %P tag and the %pb tag were merged. There is a forum thread about the changes
01:24:23Riyonukthink you could link me?
01:24:48scorche|shshe would have to do the same searching that you would have to do, likely
01:25:04RiyonukSearch for what though? %P doesn't bring results
01:25:39pixelmasomewhere in the WPS subforums, it's probably still on the first page of it
01:26:33kushal_12_27_200krazykit, I experienced the same on my Intel Macbook before
01:26:45RiyonukOh, I found it :P
01:27:01krazykitkushal_12_27_200, if osx has something like dmesg, you could plug it in and see if it gets detected at all
01:28:35pixelmalinuxstb: r18833 (right before n1s' changes plays fine), I'm compiling r18834 right now
01:29:17Riyonukpixelma, so you said earlier playing mp3 crashes for you too? So where do I get a copy that doesn't?
01:29:48kushal_12_27_200krazykit, call me crazy but it worked on the other port in the windows computer. The only thing I did different was that I held the left key on sansa until OF was fully loaded.
01:30:06pixelmaRiyonuk: where do you have the sim from?">
01:30:19RiyonukFrom there
01:31:05RiyonukiPod video, this one ->">
01:31:41kushal_12_27_200not sure why it calls it My Computer\Sansa c250\FIXED SD
01:32:03kushal_12_27_200and daaaaaaaang
01:33:16*Riyonuk wonders if everyone ditched him >_>
01:33:41krazykitRiyonuk, please have patience. IRC doesn't stand for Instant Reply Chat :)
01:33:45pixelmaRiyonuk: you would need an older one (before revision 18834 which was committed on October 19th)
01:34:10pixelmalinuxstb: 18834 is indeed the culprit
01:34:13RiyonukOh, I keep thinking my Internet connection is going out, or you guys don't care, cause you never say brb, or something -_-
01:34:20kushal_12_27_200Windows did its thing again. (note the icon for sansa)
01:34:53RiyonukPeople said stuff? ROFL, is that mIRC?
01:35:15krazykitRiyonuk, please stay on topic.
01:35:22linuxstbpixelma: Thanks for confirming that. It's odd though, as the mp3 codec doesn't use the malloc buffer...
01:35:24krazykitkushal_12_27_200, looks like your sansa is in MTP mode
01:35:33RiyonukSorry, pixelma, where might I obtain an older release?
01:35:40kushal_12_27_200ok, should I change it to autodetect?
01:35:59krazykitkushal_12_27_200, i thought sansas used MTP and MSC in the settings.
01:36:11pixelmalinuxstb: seems I can't play any music
01:36:14 Quit Thundercloud (Remote closed the connection)
01:36:26krazykitbut i suspect the information is in the manual or the wiki regarding the c200 specifically.
01:37:58linuxstbRiyonuk: You probably need to compile it yourself - I don't know of anywhere there provides old simulators.
01:38:12linuxstbRiyonuk: Or wait until the bug is fixed.
01:38:21kushal_12_27_200if you meant MTP = autodetect, it is not in that mode
01:40:24Riyonuklinuxstb: Okay, thanks. I'll stick to manually transferring to my iPod :)
01:43:04RiyonukOkay, I reformatted it, and re-installed the stable version of rockbox. I fixed the .wps %P tag, and everything looks fine. I still get "Divide By Zero" error when I play a song. Something's wrong..
01:43:46kushal_12_27_200Am I supposed to just remove the cable? :( and
01:44:02 Quit kushal_12_27_200 (" ajax IRC Client")
01:47:52 Quit barrywardell ()
01:48:16pixelmalinuxstb: ogg doesn't play either
01:53:40 Join Tetracomm [0] (n=nicholas@
01:57:18RiyonukCan I get the battery meter a percent, on the main window? I'm talking about what you see when you turn it on, it
01:57:25Riyonuk*it's usally in the top left
01:58:15 Quit krazykit ("Connection reset by beer")
01:59:17pixelmayes, there's a setting for it (somewhere in the display - statusbar settings I believe). I'd suggest consulting the manual for details... ;)
01:59:26 Part pixelma
01:59:56 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
02:00:50*linuxstb can't see what's wrong with the sim and goes to bed
02:01:37RiyonukThrough trial and error, I've found the culprit, it's this line
02:01:42RiyonukWhat exactly is wrong with it?
02:02:37linuxstbRiyonuk: This page documents the WPS syntax -
02:02:42 Quit Tetracomm ("Visit:")
02:02:58 Part AJCantos
02:03:05RiyonukYes, I'm on that page
02:03:18RiyonukLooks fine to me
02:06:20linuxstbYour line is missing the "filename.bmp"
02:06:32RiyonukWell, I have that, just was showing you an example
02:06:46linuxstbSo what exactly is your line?
02:06:51Riyonukthat :)
02:07:18linuxstbAnd what is the error message?
02:07:49 Join tonyspeaks [0] (
02:07:51RiyonukIt freezes, and says "Cant divide by zero" in the top left
02:11:13tonyspeaksHi. Domonoky asked me to find out something for him about a problematic filename for making .talk clips. Anyone know if he is likely to be here soon tonight?
02:12:04pixelma2Riyonuk: you gave the progressbar a length of 0, look at your parameters - the meaning of the numbers (and their order) has changed too. I'd still recommend searching that forum thread. Interesting error though
02:12:24RiyonukThat's what the original author did, I don't understand it either :P
02:13:06linuxstbtonyspeaks: Probably not - it's 2.12am in his timezone.
02:13:34tonyspeaksoh, well, I'll see him tomorrow then. Thanks a bunch.
02:13:52linuxstbtonyspeaks: But if you just mention the problem now, he may read the IRC logs (or I or others will point him to them)
02:14:52RiyonukOh, my theme is almost perfect :)
02:15:25tonyspeaksoh ok. He will remembver that there was a problem file that was generating an encoding error for talk clip generation with the rock box utility, and I found it it was a file with no name and only an extension thatt is located in the .rockbox folder. When I tried generating talk clips for that folder, that's where I got the error.
02:17:46linuxstbAh, so a file called something like ".rockbox/.mp3" ?
02:18:13tonyspeaksyeah, some thing like that. it was just a .talk filename with no name but only an extension.
02:19:17tonyspeaksI thought it was weird, but I wonder if the utility can be updated to bypass the .rockbox folder entirely, since it wouldn't be something that many people would access, or that maybe we can have a choice with a check box to have it generate talk clips for the .rockbox folder.
02:19:25linuxstbSo did you delete that file, and then it worked?
02:20:34tonyspeaksno. I had already generated the talk clips I needed, so I was only going through the other folders in the player to see which one would cause the error. That would be a good idea to reallyy isolate any other problem file names.
02:22:08 Join kkurbjun [0] (
02:22:21 Quit kachna (Read error: 104 (Connection reset by peer))
02:22:25linuxstbJust so domonoky can try and recreate it, what was the exact name of the file causing problems? ".rockbox/.talk" ?
02:22:29 Join kachna [0] (
02:23:46tonyspeaksit was an encoding error in my temp directory on my local harddrive. IT said that the encoding of c:\user\blahblahblah\temp\.talk failed.
02:24:15tonyspeaksso I don't know where in the rock box folder it was. It has quite a few subfolders.
02:25:06tonyspeaksNot sure if they were forward slashes or backslashes now.
02:26:59linuxstbOK, thanks.
02:27:35tonyspeaksYou're welcome.
02:31:01tonyspeaksGood night and thanks again.
02:31:13 Quit tonyspeaks ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
02:31:14 Quit jhulst (Remote closed the connection)
02:32:36 Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]")
02:36:18 Quit Riyonuk ()
02:52:40 Quit einhirn (Read error: 54 (Connection reset by peer))
02:54:07 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
03:03:12***Saving seen data "./dancer.seen"
03:25:07 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
03:32:25 Quit PaulJam (".")
03:32:44 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
03:34:54 Join toffe82 [0] (
03:53:10 Join tarelerul1 [0] (
03:53:51tarelerul1I have the gigabeat f and I was wondering how good it was under rockbox? Any of you use it
03:55:06advcomp2019it is one of the better player from what i heard
03:56:02tarelerul1The part I wonder about the most is does it use controls well. So I can surf the play easy
03:57:04advcomp2019i do not have one so i am not sure
03:59:12tarelerul1What is like on your player ?
04:00:15advcomp2019i have a sansa e280R.. you could try it on your player to see how good it is
04:00:45cool_walking_tarelerul1: Obviously this channel is full of people who like Rockbox. If you don't want to install it, you can test the simulator on your PC:">
04:00:46tarelerul1I like to look before leep so to speak.
04:02:39cool_walking_Taking a look at the manual (linked on the left of every page on will give you an idea of the features.
04:03:18tarelerul1Do any of you know how to import the songs from gigabeat? It has the cover art and tags .
04:04:35advcomp2019you can put the music anywhere.. you can read the AlbumArt wiki too
04:05:09 Join Riyonuk [0] (
04:05:38RiyonukIs there an updated version of the custom config file wiki?
04:06:25RiyonukI don't see "selector type: bar (color)", or "line selector start color: 00EB9C" anywhere. Did you guys purposely leave those out? Also, is it possible to have the selector bar transparent? As opposed to the solid color, or gradient? And is the gradient limited to only 2 colors?
04:06:40tarelerul1 The fact it encrypted the songs is lame. Even ipod don't do that. I am thinking about putting it on the gigabeat. Plus if it makes better play I am down. I hate the fact I need the software to use it .
04:09:16 Join purpledynasty [0] (n=62715da9@gateway/web/cgi-irc/
04:09:32 Join m0f0x [0] (
04:09:54 Join ameyer [0] (
04:10:33tarelerul1Does rockbox format the device ? I have use gigabeat as portable hard drive to and I have 20 movie just setting on it.
04:10:34advcomp2019tarelerul1, encrypted songs? they only songs with encryption is the DRM songs and rockbox will not play them
04:11:32purpledynastyHello everyone - this is my first time here - I've been using Rockbox for about a month on my e260, and joined the Rockbox email list about a week ago.
04:11:35 Quit synergist (Read error: 60 (Operation timed out))
04:12:32advcomp2019tarelerul1, as far as i know, it will not but since i do not have a gigabeat, i am not 100% sure on it
04:12:57advcomp2019purpledynasty, hello then
04:12:58purpledynastyIs there any reason why the database would keep corrupting where multiple copies of every track from certain albums (CD's) show up? Could it be happening because I've been updating the firmware to the nightly build almost daily?
04:13:55ameyerthe nightly build should be fairly stable.
04:14:30purpledynastyWhy does that keep happening? It forces me to initialize, and THAT causes me to lose runtime data (times played, rating, etc)
04:14:55tarelerul1advcomp2019: the reason I think it encrypted the songs is I renamed them to somet.mp3 and they did not play. They are named some.mp3.seg
04:16:01 Join synergist [0] (
04:16:08advcomp2019tarelerul1, you do not have to that with rockbox then
04:16:46tarelerul1Does it format the device it is put on ? on any of them
04:17:18 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
04:17:38advcomp2019as far as i know, no but i do not have gigabeat tho
04:18:18tarelerul1That is good. I wanted to keep the movies on it. and have system for my songs that was open.
04:18:29saratogai don't know what a seq file is, but if its not an mp3, and you renamed it mp3, then the database is unlikely ot be happy about it
04:18:40cool_walking_I don't have a Gigabeat F, but I have 3 types of Rockbox targets, including a Gigabeat S, and none are formatted by the install.
04:19:08 Join miepchen^schlaf_ [0] (
04:19:12cool_walking_The manual doesn't seem to mention anything about formatting in the installation process. I'm sure it would if that were the case.
04:20:16tarelerul1I would like to say the work you guys do is great. Opening up a device like the ipod is great idea. I mean most play don't support any other audio or video formats . most do ogg audio ,but not video format.
04:20:27saratogai think the only targets that need to be formated are HFS+ ipods
04:21:35tarelerul1thanks for being so helpful. I am sure what I asked is some where in the doc ,but sometimes find such is not easy at all.
04:21:41purpledynastyIf an HFS+ iPod is formatted - can it dual-boot and mount on a Mac after reformatting?
04:21:56saratogasure macos reads FAT32 fine
04:22:10cool_walking_purpledynasty: yes, but apparently the Mac can take a little longer recognising FAT32 partitions than HFS+
04:22:21purpledynastyOK - that means it will work in disk mode - but how about iTunes?
04:22:23tarelerul1Well, I have gigabeat and i am running Ubuntu so and it is formated to fat32 I think
04:23:01purpledynastyWhenever I've connected a Windows formatted iPod to my Macs it has asked to reformat first
04:23:22RiyonukIm curious on opinions, why would I need a txt viewer? Or image viewer? XD
04:23:34ameyerRiyonuk: why not?
04:24:02ameyeralthough, on a b&w target, it becomes "why not" and "why?"
04:24:02cool_walking_Riyonuk: I thought those were some of the least surprising plugins..
04:24:22tarelerul1Does it just do basic text file or something on the order of windows doc or open officess' formats?
04:24:37cool_walking_Just plain text.
04:24:57*ameyer assumes there have been some members of the "live music taping" community who have been involved in the development of rockbox
04:24:59cool_walking_You can open .doc, etc. files in the text viewer and probably be able to understand some of it..
04:25:21ameyerwhere stuff is circulated untagged with info in a text file
04:26:09RiyonukNo, I mean, what are some scenarios in which they would be useful? Just curious
04:26:20RiyonukPersonally, I don't see any reason for them, just wanted to know :)
04:26:21cool_walking_For keeping a todo list or something
04:26:27cool_walking_and for reading ebooks
04:27:18RiyonukAren't ebooks usually .pdf?
04:27:45cool_walking_ebooks come in a variety of formats. They can all be converted to .txt.
04:28:28RiyonukInteresting, I might see myself using it for grocery lists, or trig functions XD
04:28:41tarelerul1What is the database like for movies? Can you orginize them by tv show ,season , genre ect
04:29:14cool_walking_The database doesn't work with video. Video in Rockbox is relatively new.
04:29:37RiyonukUgh, I can't seem to find the volume .wps tag, anyone know? It's not %pv or %Sp
04:30:55tarelerul1When add cover art do you just add picture to the directory of say folder with the songs from album ?
04:32:32 Join Makuseru [0] (
04:32:41 Quit m0f0x ("dormir")
04:33:31purpledynastyget rockaa_0.6.jar a java program for album art - it runs on everything and automates getting album art for rockbox.
04:33:33RiyonukCan I force the volume to not show negatives? Thats weird
04:33:37 Quit perrikwp (" ajax IRC Client")
04:34:28 Quit miepchen^schlaf (Connection timed out)
04:35:47saratogaRiyonuk: typically only negative volume values are used, positive generally means you're clipping
04:36:22RiyonukI wish it would be positive, and like go into red if it goes over 30 or something
04:36:30purpledynastyclipping = TOO LOUD
04:36:56Riyonuklol :P
04:37:09LloreanNo, clipping = "louder than can be represented by the numbers used, so information is lost and you get a weird audible effect"
04:37:36saratoga"red" in this case would be any positive value
04:38:03cool_walking_Riyonuk: Never made a WPS before, but I think you can do that with %?pv<0|1|2|3|4|5|6|7|8|9|10|11|12|13>
04:38:10purpledynastyand the pre-digital term, when needles pointed to VU dials - it was PINNING, so clipping = pinning
04:38:24saratogaclipping is not a digital term
04:39:00purpledynastyYes, but you never hear pinning the dials any more
04:40:38purpledynastyActually, I believe I used to use the term "I just blew out my amp" :)
04:40:42saratogaI don't think pinning means anything
04:43:09cool_walking_According to urbandictionary, 'childs way of saying "spinning"'.
04:44:29purpledynastyAccording to this, the term pegging and pinning of a VU meter are synonymous and I described them well:,M1
04:45:10purpledynastyGoogle is my friend too
04:45:24purpledynastygoodnight everyone
04:45:25 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:45:29 Quit purpledynasty ("CGI:IRC")
04:45:51 Join sarixe [0] (
04:46:21saratogathat link makes it clear that pinning is not the same as clipping
04:51:07tarelerul1How good is rockbox about playing videos ? Can it do stuff like subtitles and mulit-audio track for anime
04:52:08ameyerthe #2 most-used link in this channel ever:
04:52:20cool_walking_No multi-audio. There's a patch in the tracker for subtitles, I haven't used it though.
04:52:24ameyeror, as a short answer, "not particularly"
04:54:20tarelerul1Well, if it show videos period is ok with me. That would all be nice ,but I can live out it.
04:54:38tarelerul1I don't think the ipod supports such features .
04:54:39ameyeralso, I believe it's theoretically possible to encode a video with pre-rendered subtitles
04:55:03tarelerul1hardsubs is what you mean
04:55:21ameyerstill, /me shudders at the thought of subtitles on his e260, e280, or cfmodded ipod mini
04:55:29ameyertarelerul1: I believe so
04:55:45 Join blkhawk- [0] (
04:56:08tarelerul1Mulit-audio in movie would be hot. I mean I have some anime that has dual audio .
04:56:23 Quit blkhawk (Read error: 60 (Operation timed out))
04:56:26LloreanSince you have to transcode the video to fit the screen, picking a single audio track (since you'll usually only ever have one listener) and hardsubbing aren't really a significant problem.
04:57:15tarelerul1You mean 320x240 for the video . that is cool
04:57:20 Nick blkhawk- is now known as blkhawk (
04:57:40tarelerul1On player with bigger screen subtitles might be cool.
04:57:52LloreanAs I said though, hardsubbing is an option
04:58:01LloreanThere's no real reason not to do it if you aren't going to be using the file elsewhere.
05:00:23tarelerul1I am not trying to say it should be add . I am just wondering what rockbox can do that is all. So far it sound pretty nice for songs
05:00:41Lloreantarelerul1: To find out everything it can do, just skim the manual.
05:02:12saratogalooks like someone on ABI corrupted their e200 by trying to copy a large file, presumably due to the SD controller bug
05:02:28saratogacopy within rockbox from the SD card that is
05:03:14***Saving seen data "./dancer.seen"
05:04:24 Quit Riyonuk ()
05:04:36tarelerul1Is there any device that has good linux support that does all the stuff that the video ipods or ipod nano black 8g ? I lost my ipod and I was what I should get next to replace it.
05:04:58Lloreantarelerul1: This channel is specifically about Rockbox, not general device questions.
05:05:28tarelerul1I just thought one of you might know of one.
05:07:01saratogatheres a community channel if you want to ask about other things
05:07:15ameyersaratoga: wonder-freaking-ful. I guess it's not that surprising, though.
05:17:32 Join esthar [0] (
05:26:25 Quit MarcGuay ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
05:28:44 Join Seed [0] (
05:29:20 Quit Makuseru (Read error: 104 (Connection reset by peer))
05:34:23 Quit saratoga ("CGI:IRC (EOF)")
05:42:23 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:53:45 Quit BHSPitMonkey (Remote closed the connection)
06:04:10 Join HBK1 [0] (
06:04:10 Quit HBK- (Read error: 104 (Connection reset by peer))
06:04:44 Nick HBK1 is now known as HBK (
06:06:49 Quit esthar (Read error: 110 (Connection timed out))
06:09:46 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:11:41 Quit Zarggg ()
06:15:01 Join esthar [0] (
06:29:57 Part tarelerul1
06:47:43 Quit jhulst (Read error: 60 (Operation timed out))
06:49:20 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
06:51:20 Quit XavierGr ()
06:53:06 Quit tim__b (Remote closed the connection)
06:53:27 Join tim__b [0] (
06:57:35 Join pierre- [0] (
07:03:18***Saving seen data "./dancer.seen"
07:11:59 Quit scorche (Nick collision from services.)
07:12:51 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
07:31:23 Nick fxb__ is now known as fxb (
07:38:16 Nick fxb is now known as fxb__ (
07:41:59 Quit pierre- (Read error: 60 (Operation timed out))
07:50:40 Quit scorche (Read error: 104 (Connection reset by peer))
07:51:43 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
07:53:48 Quit BHSPitMonkey (Remote closed the connection)
07:55:21 Quit miepchen^schlaf_ ()
08:04:08 Quit toffe82 ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
08:06:43 Join kronflux [0] (
08:12:03 Quit kronflux ("Leaving")
08:13:11 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
08:13:12 Quit reacocard (Read error: 110 (Connection timed out))
08:13:34 Quit BigBambi (Remote closed the connection)
08:15:32 Join reacocard [0] (n=reacocar@DHCP-61-115.ST.HMC.Edu)
08:23:00 Join bertrik [0] (
08:35:20 Quit Zom (Read error: 60 (Operation timed out))
08:36:07 Join Zom [0] (
08:38:11 Join culture [0] (
08:49:50 Quit bertrik (Remote closed the connection)
08:50:54 Quit Zom (Read error: 60 (Operation timed out))
08:50:58 Join Zom [0] (
08:53:51 Join ender` [0] (
08:59:07 Join Rob2222 [0] (
09:00:38 Join BigBambi [0] (i=86ceaf40@rockbox/staff/BigBambi)
09:02:19 Join petur [50] (n=petur@rockbox/developer/petur)
09:03:23***Saving seen data "./dancer.seen"
09:05:34 Quit culture (Read error: 110 (Connection timed out))
09:09:36 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
09:10:01 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
09:10:45 Quit kachna (Read error: 113 (No route to host))
09:16:35 Join avis [0] (n=ident@pdpc/supporter/student/avis)
09:16:46 Quit Rob2223 (Read error: 110 (Connection timed out))
09:22:06 Join gromit`` [0] (
09:29:39 Quit avis ()
09:29:59 Quit gromit` (Read error: 110 (Connection timed out))
09:31:26 Join Nibbl [0] (
09:33:24 Join Thundercloud [0] (
09:40:41 Quit pixelma2 ("-")
09:40:51 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
09:46:55 Join BigBambi_L [0] (i=86ceaf34@gateway/web/ajax/
09:49:16 Join avis [0] (n=ident@pdpc/supporter/student/avis)
09:52:53 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:53:28 Quit jhulst (Remote closed the connection)
10:12:41 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
10:19:12 Quit avis ()
10:23:30 Quit n17ikh|Lappy ()
10:29:23 Quit Thundercloud (Remote closed the connection)
10:50:31 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
10:50:41 Quit linuxstb (Nick collision from services.)
10:50:45 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
10:51:24 Join tvelocity [0] (n=tony@
10:53:10linuxstbpetur: Can you remember if there was a reason you used "#ifndef SIMULATOR" when you added the ata spindown handling to the jpeg viewer? (r11299 praises you for that change ;) )
10:53:49peturerrr... no
10:54:55 Quit Slack ("Ex-Chat")
10:55:00*petur busy right now
10:57:26linuxstbpetur: OK, don't worry - I'll test things and commit if it works.
10:58:40 Quit cool_walking_ (" ajax IRC Client")
11:00:03peturis there a reason to call ata_spindown in the simulator?
11:02:49 Quit tvelocity ("Αποχώρησε")
11:03:25***Saving seen data "./dancer.seen"
11:04:58linuxstbTo remove as many #ifndef SIMULATOR lines as possible. Ideally there shouldn't be any at all in apps/ code.
11:14:29 Quit lordpil (Read error: 113 (No route to host))
11:15:18*petur would also rather have stubs than #ifndef
11:15:39*B4gder agrees
11:15:49B4gderthe more we can hide simulator-stuff, the better
11:19:34linuxstbYes, stubs exist for the ata_spindown family of functions - and they seemed to exist at the time the spindown stuff was added to the jpeg viewer, which is why I asked if there was another issue.
11:21:25peturI guess I didn't check very well and assumed it wouldn't exist?
11:22:27*petur checks to see if there is a corner free and walks over
11:24:36 Join Rob2223 [0] (
11:25:44 Quit Rob2222 (Connection reset by peer)
11:30:58linuxstbamiconn, gevaerts: A few days ago, you discussed the use of the plugin api pointer in the plugin lib. Would it be a bad solution to simply introduce the requirement that all plugins using the plugin lib must declare a global "const struct plugin_api rb;", which is then used by the library? If plugins fail to do it, then the linker should complain.
11:33:05 Join JdGordon [0] (
11:36:16 Join goffa [0] (n=goffa@
11:45:15pixelmalinuxstb: so, you do not have an idea what in n1s' codec ram changes could have broken playback in a windows sim? Wonder who I should ask...
11:45:41 Quit Nico_P (Read error: 60 (Operation timed out))
11:46:04linuxstbpixelma: I looked last night, but couldn't see any reason why it would break. I think we need to ask everyone to look...
11:46:46B4gdercould it be related to the size of the codec and the memory buffer it gets loaded into now?
11:47:04B4gderI mean the alloc part perhaps isn't big enough?
11:47:15 Quit goffa_ (Read error: 110 (Connection timed out))
11:47:37linuxstbNot all codecs use the malloc buffer - and even codecs that don't touch it (like mp3) break.
11:47:49pixelmais there something I could check or try to find out more?
11:47:51B4gderso all codecs break?
11:48:04 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:48:23pixelmaB4gder: yes, looks like it. I tried mp3, wma, ogg and the more exotic ones adx and mod
11:48:44linuxstbB4gder: And it works fine for me (on Linux)...
11:48:47JdGordonNico_P: hey, did you end up looking at that patch?
11:49:05Nico_PJdGordon: no, sorry I didn't
11:49:22Nico_PI don't have much rockbox time ATM
11:50:12JdGordonok... its a 1 line change... can you have a quick look and decide if it should go in?
11:53:00Nico_Plooking now
11:54:16Nico_Pwell it does look perfectly harmless, so if it fixes the problem, I don't see why it shouldn't go in
11:54:52JdGordonjust that it will hide the real issue which i guess is something in pcmbuf goes wierd when seeking?
11:55:07*JdGordon back in 10
11:55:19 Quit perrikwp (" ajax IRC Client")
11:56:54JdGordonquick 10 :p
11:56:56Nico_PI don't know the pcm code well enough to comment
11:57:11Nico_Pactually, I don't know it at all ;)
11:57:39JdGordonanyone know the reasons we can do voice while paused? is it because they are mixed in pcmbuf too early?
11:58:06Nico_Pdid you mean "can't"?
11:58:41JdGordonyes :p
11:59:35Nico_PAFAIK it's because pause is done by pausing PCM
11:59:58Nico_Pso yes, I guess it's because they're mixed in too early
12:00:30JdGordonany ideas how expensive mixing pcm is?
12:00:59linuxstbI think it's also because there is only one pcm buf. So if it's full of decoded audio, then you can't play other things. IIRC, this has been discussed a few times in the past, with the conclusion that we need multiple PCM buffers, and just-in-time mixing.
12:01:34JdGordonI was thinking about it during the last few days and I want to give it a try...
12:02:18linuxstbTry searching the IRC logs (possible for more than 1 or 2 years ago...) - I'm sure you'll find a discussion with jhMikeS, amiconn, preglow and others.
12:03:41 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
12:03:43JdGordonis mixing expensive?
12:04:02JdGordonor is it just lots of adding?
12:05:24linuxstbNo, I don't think it's expensive. There may even have been the suggestion that it was done in the ISR
12:05:56linuxstbBut as I said, best to find the previous discussions.
12:20:11 Join moos [0] (
12:20:58 Quit JdGordon ("Konversation terminated!")
12:23:03 Join JdGordon [0] (
12:34:09 Join tvelocity [0] (n=tony@
12:34:33 Join Acky [0] (
12:41:49 Quit nuonguy ("This computer has gone to sleep")
12:43:05 Nick fxb__ is now known as fxb (
12:48:31 Join kachna [0] (
12:51:20 Quit Acksaw (Connection timed out)
13:03:26***Saving seen data "./dancer.seen"
13:23:22 Join nplus [0] (n=nplus@141.25.Globcom.Net)
13:23:41 Quit JdGordon (Read error: 104 (Connection reset by peer))
13:24:02 Quit kachna ("Konversation terminated!")
13:26:30 Join JdGordon [0] (
13:32:38 Join LambdaCalculus37 [0] (
13:53:50 Quit LambdaCalculus37 ("Ka-chunka")
13:54:10 Join LambdaCalculus37 [0] (
13:54:13 Quit LambdaCalculus37 (Remote closed the connection)
13:55:05 Join kushal_12_27_200 [0] (n=kushal@
14:00:16 Part LinusN
14:03:27 Join Arathis [0] (
14:03:48pixelmalinuxstb: umm, a reconfigure seems to have fixed it... I think I did a "make clean" before, no rerun of configure though...
14:04:25pixelmabut it didn't give me any warnings or errors during th compile
14:04:43pixelmagotta go though, back later
14:04:54 Part pixelma
14:05:11 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
14:08:14linuxstbpixelma2: Hmm, I forgot that that could be a problem. I wonder if rasher's sims are built in a similar way - rasher?
14:12:24 Join LambdaCalculus37 [0] (i=44a04303@gateway/web/ajax/
14:18:57 Part Arathis ("Bye, bye")
14:23:19 Quit amiconn (Nick collision from services.)
14:23:25 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
14:26:23 Join PaulJam [0] (
14:39:24 Quit pixelma2 (Nick collision from services.)
14:39:32 Join pixelma2_ [0] (n=marianne@rockbox/staff/pixelma)
14:39:34 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
14:46:46 Quit moos (Read error: 104 (Connection reset by peer))
14:54:13 Join moos [0] (
14:58:06 Join J-23 [0] (
15:03:27***Saving seen data "./dancer.seen"
15:07:45 Join robin0800 [0] (
15:20:50 Join Schmogel [0] (
15:21:44 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
15:27:31 Quit robin0800 (Connection timed out)
15:33:20JdGordonLambdaCalculus37: you pinged me a couple of days ago?
15:35:08preglowlinuxstb: thunk anymore on how to use the auto-generated codec memory info?
15:36:04linuxstbpreglow: You mean how to display it usefully?
15:36:53preglowwe will probably just have to link to auto-updated files in some fashion, updating the wiki all the time will prove... boring
15:37:24preglowhmm, perhaps having the perl tool scan all map files, outputting html....
15:38:39B4gderisn't the problem still that the malloc use in some codecs still make those numbers a bit "random"?
15:38:56preglowB4gder: oh yeah, but target is that malloc will be gone anyway
15:39:01preglowand most codecs don't use malloc
15:39:04preglowonly two major ones are left
15:39:17linuxstbMy idea was to have another file with a textual description of the malloc use of that codec, which the script will include in the output.
15:39:59B4gderone obvious use is of course to trim the codec buffer according to the largest codec + malloc case
15:40:28B4gderbut that isn't about presentation of the data...
15:40:35preglowif we malloc use is static enough across different files, sure
15:40:37linuxstbProblem is (I think) that some malloc usage is unbounded - Rockbox will currently fail on some files due to lack of memory
15:40:37preglowbut yeah :)
15:40:58preglowlinuxstb: oh? format?
15:41:19LambdaCalculus37JdGordon: Yes. I posted an OS X version of Sansapatcher to FS #9369 a couple of days ago. Want to give it a test run?
15:41:30linuxstbmp4 seektables are one thing that come to mind - I think some very long (and low bitrate) AAC files fail
15:41:40preglowLambdaCalculus37: you don't have native os x rockbox compiling going, do you? :>
15:41:43linuxstbBut maybe (hopefully) I'm wrong...
15:41:51preglowlinuxstb: ah, so no codec stuff
15:42:04LambdaCalculus37preglow: I do.
15:42:06preglowlinuxstb: seek tables are low priority, really, we should dump them if they're too long, so not a big problem
15:42:13JdGordonLambdaCalculus37: yeah, tomorow though... I dont have a usb cable going from it to my room :p
15:42:19preglowLambdaCalculus37: stellar! how'd you go about that? i can't even get cross compilers to build
15:42:22linuxstbpreglow: No, I don't know much about the insides of codecs - I just deal with the higher-level parts of codecs...
15:42:52preglowlinuxstb: vorbis, for example, could possibly go ahead and explode malloc buffer with a very large custom codebook
15:42:54linuxstbpreglow: MP4 seektables more than that - they're used when decoding as well (they store the lengths of frames)
15:43:00preglowhmm, no, probably doesn't have indices high enough
15:43:08preglowlinuxstb: bah...
15:43:16preglownow what the hell kind of a design is that
15:43:24LambdaCalculus37preglow: I just used in the tools folder.
15:43:31linuxstbpreglow: But someone needs to dig into the two mp4 codecs and investigate and see if that can be changed.
15:43:34preglowLambdaCalculus37: interesting... and this is stock leopard?
15:43:36LambdaCalculus37It built the entire cross compiler and environment for me.
15:43:41LambdaCalculus37preglow: Tiger.
15:43:47preglowthat might be it, then
15:44:10LambdaCalculus37preglow: Why, is Leopard giving you troubles?
15:44:19preglowLambdaCalculus37: well, like i said, can't get cross compilers to build :)
15:44:31preglowso i currently do all rockbox stuff in vmware, which is SLOOOOW with usb
15:44:33preglowdamn annoying
15:45:03GodEaterpreglow: is VMWare still USB1.1 only ?
15:45:09preglowmight be...
15:45:12LambdaCalculus37preglow: If it helps any, I also have XCode installed and also installed stuff like make, gcc, and others via Fink.
15:45:43preglownah, seems fusion does usb 2
15:45:49preglowLambdaCalculus37: yeah, i have those too
15:45:55JdGordonpreglow: ive got leapord and didnt have any trouble getting thr cross compilers going... just installed xcode and rand the build scri[t
15:46:09preglowJdGordon: well shit, i'll give it a go again next time i do rockbox coding, then
15:46:13LambdaCalculus37Same setup here on Tiger.
15:46:30preglowi won't be pissed if i don't have to go through vmware
15:47:04preglowand the mac is so damned sweet i can't be bothered to turn on the pc anymore...
15:47:06preglowit's noisy :P
15:47:25*LambdaCalculus37 nudges preglow back into on-topic talk :)
15:47:28JdGordonit also not a mac though...
15:47:52preglowLambdaCalculus37: talking about beer this time of day???
15:48:22LambdaCalculus37preglow: Hehe... :)
15:49:10 Nick bmbl is now known as BMBL (n=Miranda@unaffiliated/bmbl)
15:49:42preglowwell, anywho
15:55:35 Join lasser [0] (
15:55:43 Join funman [0] (
15:57:31 Quit Seed ("cu, Andre")
16:00:03rasherlinuxstb: My sims are built using mingw32 on Linux
16:00:49rasherlinuxstb: also, I do a clean build each time
16:01:53 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
16:02:07 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
16:04:48funmanI'm getting better results from the Clip's (supposed) SDRAM : with 8 or 16 bits access, 1/8 of the memory can be read/written reliabily
16:05:39funmanwhatever size I define (I was afraid it was 2 Mbits wide, but it's theorically 16Mbits : 2Mbytes)
16:16:14funmanafter power up the ram is filled with 0xffff
16:18:32 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
16:18:42 Join culture [0] (
16:18:57funmanafter I have written [i] = i; with 16 bits access, I read [i] == (i%16)/2;
16:20:34 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
16:23:05 Quit linuxstb (Read error: 110 (Connection timed out))
16:33:57funman how do you understand 'configure the static memory configuration' ?
16:34:07funmanIn: what is static, the memory or the configuration ?
16:36:58funmanfollowing registers are named 'static memory xxx' so I suppose it's the memory => I can ignore them for SDRAM
16:39:03B4gderyes, sdram is dynamic
16:41:39B4gderdo you know the sdram setup from the OF?
16:42:21funmanyes, and I have the datasheet of the memory controller (ARM PL172)
16:42:43B4gderok, so the refresh and RAS/CAS stuff etc should be correct?
16:43:05funmanhow ever due to how the OF uses zillions of stack variables, transmitted over zilliards of functions, I may have wrongly determined some settings
16:43:20funmanright, "should"
16:43:29B4gderI see
16:43:45funmanas far as i understand, some parameters are even read from uninitialized stack area
16:45:43funmanI was hoping the famous sansav2 reverse engineer would come back to help me ;)
16:46:16 Join ghen [0] (
16:46:57ghensvn up'ed from a few days ago, now getting the following build error on NetBSD:
16:46:58ghenMAKE in libspeex for rockbox core
16:46:58ghengmake[2]: *** No rule to make target `/data/cvs/rockbox/firmware/export/id3.h', needed by `/scratch/rockbox/build/ipodvideo/apps/libspeex-rockbox/bits.o'. Stop.
16:47:02ghengmake[1]: *** [libspeex-rockbox] Error 2
16:47:06funmanghen: make clean
16:47:19B4gderghen: and possibly reconfigure
16:48:00 Quit GodEater (" ajax IRC Client")
16:49:03ghenmake clean was sufficient
16:49:16 Quit BigBambi_L (" ajax IRC Client")
16:49:20funmanmy kingdom for a jtag connection to a sansav2 !
16:49:53 Quit perrikwp (" ajax IRC Client")
16:50:08ghen(only Makefile diff after reconfigure was absolute paths for the tools)
16:50:10 Quit kushal_12_27_200 ("This computer has gone to sleep")
16:50:23 Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]")
16:50:52B4gderright, they're only two sensible things to try when things break like that
16:51:35ghenright, my bad not to try that first, sorry ;)
16:51:38funmanghen: problem was in dependancies file not being regenerated
16:52:58funmanwhen e200v2's LCD is working, I'll have some test requests for e200v2 owners ..
16:53:37 Quit Unhelpful (Read error: 113 (No route to host))
16:55:30funmanlinuxstb: what's missing in e200v2 lcd code ? I see 'todo: more' in comment of ams3525_dbop_init() but it's not very meaningful
16:57:58 Quit Zagor ("Client exiting")
17:03:28***Saving seen data "./dancer.seen"
17:03:56 Quit kkurbjun (Remote closed the connection)
17:05:46 Join pierre- [0] (
17:09:16 Join n1s [0] (n=nils@rockbox/developer/n1s)
17:12:04 Join m0f0x [0] (
17:15:04 Part B4gder
17:17:27 Quit BigBambi (" ajax IRC Client")
17:32:20rasherI wonder what made Riyonuk's sims crash. I downloaded an e200 sim - it played MP3 just fine
17:34:33 Join Xerion [0] (
17:42:38 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
17:44:26 Join faemir [0] (
17:49:12 Join {phoenix} [0] (
17:49:21 Quit {phoenix} (Read error: 54 (Connection reset by peer))
17:52:24 Join {phoenix} [0] (
17:54:46 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
17:59:49 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:02:17 Join bughunter2 [0] (n=Jelle@
18:04:22linuxstbfunman: Re: the e200v2 lcd code - I simply haven't finished understanding everything in the OF disassembly. "TODO: More" means that in that location in the OF, there is a call to a function, and I haven't disassembled that function yet.
18:05:17linuxstbSimilarly, the function I've called "lcd_delay" is (I think) a delay function - I've just got as far as spotting that it does something with the TIMER registers.
18:07:58funmanlinuxstb: for the delay just use something random
18:08:24funmanwhen you look again at this code tell me, so I can lend you my sansav2-experimented eyes ;)
18:08:38 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:08:51linuxstbOK, but I've got no immediate plans to - I've been distracted with cleaning up Rockbox's jpeg viewer
18:11:44*domonoky just recieved his google shirt ! :-)
18:11:59*linuxstb too
18:12:49*scorche wonders if either of them are planning on posting to the mentor's list about it
18:13:25*linuxstb thinks that would just be silly
18:14:10 Quit petur ("*plop*")
18:27:02 Quit culture (Read error: 110 (Connection timed out))
18:30:31 Join miepchen^schlaf [0] (
18:30:45 Join bertrik [0] (
18:33:08bertrikfunman, I just read you made more progress on the clip sdram
18:33:57funmanbertrik: if you mean 'progress' like in 'different but still broken results', then yes ;)
18:34:02bertrikdo you know how much sdram there is? did I understand correctly that is has 2 megabytes?
18:34:41*bertrik should post/commit his as3525 codec register stuff somewhere
18:34:48 Join mf0102 [0] (
18:34:58funmanIf I trust the setting of MPMCDynamicConfig0-3 registers and what we found in the OF, yes 2Mbytes for Clip, and 8 for E200v2
18:35:13funmanbertrik: you can use the git repo as a 'draft'
18:36:08funmanbut still, the SDRAM looks pretty big on the E200, I don't know where it could be hidden on the Clip
18:36:37funmanand saratoga had some arguments why it would be hard/expensive to put SDRAM inside the SoC package
18:37:38linuxstbAh, so the E200 definitely has an external SDRAM chip?
18:38:08linuxstbThe TCC SoCs (some versions) have 2MB embedded SDRAM - and these are used in very cheap flash players.
18:38:10funmanyes, it's very visible on the photos
18:38:23bertrikI think we should stick to a single repository to avoid having to synchronise repositories. If the new code doesn't affect other targets, it should be put in rockbox svn, IMO
18:38:51funmanbertrik: I use the gitorious repo to share incomplete/broken code with other sansav2 hackers
18:39:13funmanand ignore completely any synchronization
18:39:25 Join nuonguy [0] (
18:39:28*linuxstb is also in favour of that - incomplete/broken code is fine in SVN at this early stage of a port (e.g. the e200v2 lcd driver)
18:39:34bertrikI think AMS/sandisk did a great job combining several types of electronics (analog, digital, etc.) in the SoC and 2 MB sounds doable to add
18:40:04domonokyfunman: why would it be difficult to integrate SDRAM into a SoC ? the memory structure itself shouldnt be a problem, as modern CPUs have x Mb cache, so Sd ram shouldnt be a problem..
18:40:22dionoeahi. Does USB charging work for e200v1 in rockbox?
18:40:30funmandomonoky: ask saratoga, not me (or maybe I have misunderstood him)
18:40:33*bertrik hides
18:41:18bertrikUSB charging does work in general, but there are a couple of weird issues and the code structure is not very clean
18:41:19funmanlinuxstb: incomplete/broken; like in any crap ?
18:41:27 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
18:42:02dionoeabertrik: when I look at the battery monitor and have, let say a battery level of 50% and I plug the USB, battery levels goes up 1% every 1/2 second
18:42:06dionoeakind of weird :/
18:42:15dionoeahey funman !
18:42:28bertrikcrap can be exchanged via a pastebin, incomplete but correct stuff can be put in svn, IMO
18:42:37funmanhey :)
18:43:01funmanI think a central repository permits easier synchronization
18:43:32 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
18:43:40funmanof course Someone can take bits from the git repository I use and commit them to rockbox svn
18:43:56linuxstbfunman: Speaking personally, I would prefer that you committed crap to our SVN than your own GIT repository - it's nice if everyone is looking at the same code.
18:43:59bertrikdionoea, I think the 1% per half second is a general powermanagement feature
18:44:17dionoeabertrik: I didn't expect battery charging to be that fast :)
18:44:51funmanlinuxstb: since a very few people actually work on the sansav2 port I think all useful stuff has been committed to rockbox svn already
18:45:08bertrikdionoea, the battery reading in current rockbox is incorrect (way too high during charging) so that's why it keeps incrementing
18:45:20dionoeaah ok. Thanks.
18:45:51bertrik*incorrect during charging, I should say, when not charging the battery reading is OK
18:48:04linuxstbfunman: Sure, I don't think it's a big issue at the moment, but I just have bad memories of other ports, where lots of things happened outside Rockbox SVN, resulting in a) lots of work to sync to Rockbox SVN and finally commit; and b) strange code that nobody understands because it wasn't reviewed at the time it was written.
18:48:59 Nick m0f0x is now known as colesterol_dog (
18:49:04funmanWhen some progress is made I will manage to commit into rockbox svn
18:49:51 Quit sarixe ("Ex-Chat")
18:50:22bertrikfunman, great, thanks
18:51:12funmanif i'm not dead/away of course
18:51:32linuxstbYou planning something? ;)
18:51:39linuxstbbertrik: How is the i2c stuff going?
18:51:45saratogawhich telechips part has internal SDRAM?
18:52:11linuxstbSome of the tcc77x variations. They have different combinations of embedded NOR flash, audio codec and SDRAM.
18:52:26funmanlinuxstb: yes - leave for the mountains
18:53:28linuxstbsaratoga: - the tcc772 and tcc773. The tcc773 is in the Logik DAX.
18:53:47bertriklinuxstb, it's done basically. It uses polled mode though, while the OF seems to use interrupts.
18:55:04 Join jgarvey [0] (
18:55:22bertrikAs far as I can tell, the specific i2c bus I worked on is only used to communicate to the codec part of the as3525, so I made a simple API like as3525_codec_read(index) and as3525_codec_write(index, value) that reads or writes and hides the underlying i2c
18:55:46linuxstbAny idea if the bootloader is going to need i2c?
18:55:57saratogalinuxstb: thanks hadn't seen that
18:56:05bertrika transfer typically takes about 75 microseconds, or about 30 bits @ 400 kHz
18:56:12saratogai wonder if its actually on die or if they use two seperate dies
18:56:41bertriklinuxstb, I think the bootloader does not need i2c
18:57:05funmanIIRC the i2c is initialized after the SD
18:57:26linuxstbsaratoga: No idea. But the tcc820x appears to come in variations with up to 8MB SDRAM...
18:57:56*domonoky would think that they just uses a softcore for the as, added needed parts, compiled them, and send them to some manufacturer, creating something called a asic :-)
18:58:26bertrikthe codec part contains the audio volume/levels, audio inputs, charger, supply voltages, unique id, rtc, adc
19:01:25bertrikthe code is disappointingly simple compared to how complicated it look in the OF :)
19:01:46n1spreglow: do you think it would be worthwhile to try to borrow stuff from the Tremor low-mem branch to get rid of (some of) the mallocs in our tremor codec? It does seem to use less mallocs from a quick glance.
19:02:32linuxstbn1s: I thought the lowmem branch sacrificed some speed for lower memory usage?
19:03:32***Saving seen data "./dancer.seen"
19:03:41 Join hannesd [0] (
19:03:56n1slinuxstb: i only locked at it very quickly so I don't know much of the details but it seemed to pass struct pointers around where our Tremor mallocs (for code that doesn't seem performance critical to me)
19:05:26preglown1s: anything goes. would need to be benchmarked, tho
19:05:44n1sof course
19:06:00amiconndomonoky: Re your earlier question - caches are almost always SRAM
19:06:17 Join culture [0] (
19:07:24domonokyamiconn: yes, but isnt SDRAM easier to build then SRAM ? (only on transitor needed against 6 for SRAM) .. ?
19:08:12Nico_Plinuxstb: about the jpeg decoder, do you know what kind of size it would be if added to the core?
19:08:17amiconnIt's *different* to build. It only needs one transistor so takes less space, but it needs a capacitor which SRAM doesn't need
19:09:20 Join hannesd_ [0] (
19:09:20domonokya capacitor ? isnt that just two copper lines acting as capacitor in modern designs ?
19:09:54*domonoky tries to remember what he learnt at uni... :-)
19:09:54linuxstbNico_P: Guessing based on the .o files, around 20KB on ARM.
19:10:21Nico_Pthat sounds quite OK
19:10:37linuxstbNico_P: The issue would be the ram for decoding.
19:10:57bluebrotherdomonoky: well, that would be dependent on the frequency (i.e. this is HF-stuff)
19:11:01 Join Thundercloud [0] (
19:11:22Nico_Plinuxstb: where the decoded bitmap is stored, you mean?
19:11:49linuxstbYes, and the compressed image. Although I'm guessing that's probably already on the audio buffer?
19:12:12Nico_PI would rather see the decoded bitmap in the audio buffer
19:12:38Nico_Pthere's no way to decode directly from disk?
19:13:17linuxstbI don't know, I haven't looked at the details of the decoder itself yet.
19:14:12 Quit funman ("leaving")
19:14:46*amiconn wonders why domonoky isn't in -community. This dram stuff isn't really on topic...
19:15:00*domonoky joins #rockbox-community
19:16:03domonokythanks for the pdf link, it seems uni is already too long ago... and to make it on-topic: Beer ! :-)
19:16:15Nico_PSlasheri: ping
19:16:21SlasheriNico_P: hello :)
19:16:51Nico_PSlasheri: I was teking a quick look at FS #8295 (database integration of pictureflow)... have you seen it?
19:17:00 Quit pierre- (Connection timed out)
19:17:01 Quit saratoga ("CGI:IRC (EOF)")
19:17:03SlasheriNico_P: probably not, i will check
19:17:06 Join pierre- [0] (
19:17:14amiconnNico_P: Currently the jpeg decoder doesn't decode directly from disk. I guess this could be added. You'd still have to prove the memory for storing the decompressed image, and you would need to specify the zoom level to decode
19:17:32amiconnThe decompressed image as spit out by the jpeg decoder is YUV, btw
19:17:48 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
19:18:00amiconnSlasheri: Will there ever be a fix for the empty db bug?
19:18:08 Join kachna [0] (
19:18:36 Join roolku [0] (
19:18:38Nico_Pamiconn: I guess the decoded image could go into the audio buffer (is its being YUV a problem?). specifying the zoom level to decode simply requires knowledge of the size of the JPEG and the target size, doesn't it?
19:18:45Slasheriamiconn: yes, there will.. :/
19:19:26roolkuNico_P: hi, I noticed interest in FS #8295 :)
19:19:34Slasheriamiconn: just wondering nobody has fixed it yet.. i will try to have a look at some weekend
19:19:38amiconnNico_P: The possible zoom levels are 1/8, 1/4, 1/2 and 1
19:19:57Nico_PSlasheri: it *seems* to me that the patch adds a char buffer to all DB entries, but I really don't know much about the DB. what I was thinking would be good is having a table of all cover bitmaps, and track entries could have "pointers" to entries of the covers table
19:20:05amiconnAnd the decoder could tell the resolution, so that the client could select the best match
19:20:36Nico_Proolku: hi :) yes, I'm interested
19:20:40amiconnBeing yuv means you need to do the yuv->rgb conversion in the core as well. This is not the same yuv as mpeg, btw
19:21:13amiconn(the yuv problem obviously doesn't apply to greyscale targets, there you will have to dither instead)
19:21:41Nico_Pamiconn: hmm and we'd also need to resize after doing the yuv->rgb conversion
19:21:43roolkuNico_P: at the moment the leaf node of any navigation would be a (unique) list of cover image paths
19:21:49SlasheriNico_P: indeed, that adds a (huge) string tag for every track in the db
19:21:58amiconnNico_P: Uh, why?
19:22:18SlasheriNico_P: and what worse, by default that gets loaded in ram too unless some handling is done to prevent that (like with tag_filename)
19:22:24roolkuNico_P: I intended to pass this on to the plugin which would return an index (the selection) to that list
19:22:25Nico_Pamiconn: the power of two factors will probably be a bit rough
19:22:39amiconnAh, you mean further scaling...
19:22:50amiconnWe don't even have a decent BMP scaling....
19:22:57 Join stoffel_ [0] (
19:22:58roolkuNico_P: the database tree code would then again deal with it an show a list of tracks
19:23:05SlasheriNico_P: (of course only with tagcache loaded to ram)
19:23:41Nico_Proolku: I think it would be better to make the core DB handle most of it, so we can do other cool stuff with the album covers
19:24:03Slasheriroolku: at least it would be good to avoid that tag getting loaded to ram.. or it would consume all of the ram
19:24:04Nico_PSlasheri: does the DB structure allow a table of covers?
19:24:11amiconnCool stuff isn't worth anything if the basics are broken :(
19:24:24 Quit hannesd (Connection timed out)
19:24:25 Nick hannesd_ is now known as hannesd (
19:24:26roolkuSlasheri: yes, I can see that
19:24:33Nico_Pamiconn: by basics, do you mean the DB?
19:24:36amiconnRight now, "Insert shuffled" is borked. I don't know what revision caused it though
19:24:52amiconnNo, I mean even the most basic things like playing a playlist properly
19:24:57*preglow crosses fingers and runs the cross compiler script
19:24:58Nico_Poh, right
19:25:31amiconnThis isn't the playback engine's fault, as it happens both on hwcodec and swcodec
19:25:38SlasheriNico_P: not that kind of "tables" at the moment
19:26:18amiconnTry the following: With playback stopped, hover over a folder (preferably an album folder where the track filenames start with the track numbers), and select Playlist->Insert shuffled from the context menu
19:26:29roolkuSlasheri: is this currently done with the filename ? I should look at that
19:26:37SlasheriNico_P: hmm, not sure what you meaned.. in fact all track entries are pointers to other "tables"
19:26:41amiconnGo to the WPS, then view the current playlist and write it down.
19:26:44preglowamiconn: 3.0 borken too?
19:26:54Nico_PSlasheri: the "unique" tags, is that it?
19:27:07amiconnStop Playback, and resume -> It will (almost always) resume the correct track, but in a wrong playlist position
19:27:18SlasheriNico_P: yes, like with artists for example
19:27:28Slasheriit should work the similar way
19:27:44amiconnThe order of tracks in the list stays the same when viewed as a "ring", but it's rotated so that the first track of the folder is in position 1
19:27:45Nico_PSlasheri: so each artist name is only once in the DB, and track entries have a pointer to it?
19:28:07 Join mrkiko [0] (n=IRCExplo@
19:28:13Nico_Pyes, so cover filepaths should work the same
19:28:15Slasherithat is exactly same as the current "covers" implementation seems to be
19:28:15roolkuNico_P: this is the case with the album art strings as well
19:28:25amiconnI wonder what may cause such a weird effect. I verified it on archos Player, Recorder, and iriver H180
19:28:26mrkikoHi! I'm back again... I'm again free
19:28:27Slasheriexcept those shouldn't loaded to ram
19:28:38 Quit courtc (Read error: 60 (Operation timed out))
19:28:39mrkikobye bye people and thank you for your help on rockbox life
19:28:41roolkubecause they are rather long
19:28:42preglowamiconn: can't seem to trigger it on h120 3.0
19:28:55 Quit mrkiko (Client Quit)
19:29:02amiconnSo some binchopping seems in order...
19:29:30*Nico_P should play with the patch a bit
19:29:58Slasherilike with tag_filename, tag_cover could be replaced with dircache pointers at runtime so the actual filename table doesn't need to be looked at all when db has been loaded to ram
19:29:58*amiconn wonders whether test_codec will be fixed soon
19:30:03 Join courtc [0] (
19:30:09preglowamiconn: what's wrong with it?
19:30:11Slasheri*cover table
19:30:25amiconnIt doesn't compile since the codec ram rework
19:31:17preglowthe n1s commit?
19:31:46Slasheriroolku: doing those two things would be great.. then the cover path entry would only consume 4 extra bytes at runtime when db is in ram.. The only thing needed to do would be add hooks/checks in the db ram loader and search routines
19:32:09amiconnpreglow: yes
19:32:15preglown1s: working on it, or?
19:33:08preglowgod damn it, os x still doesn't work, now with a new error
19:33:57roolkuSlasheri: I tried to find 'save' hooks, but the tagtree code is really hard to follow. There are many similar pointers to index level an position and I am never quite sure which level they relate to and which ones need to be updated. It has been a while though since I looked
19:34:10roolkumaybe I should have another go...
19:34:37roolku..and position...
19:34:42rasherpreglow: sounds like you need to apply the Power of UNIX[tm]
19:34:52Slasheriroolku: you should only have to modify tagcache.c to do that.. those aren't real hooks but just some part of the code
19:34:54preglowos x is supposed to be a unix :(
19:34:58LambdaCalculus37preglow: What's the error?
19:35:08Slasheriroolku: just follow all tag_filename checks in tagcache.c and try to do something similar with the covers
19:35:10preglowLambdaCalculus37: i'll have to pastebin this...
19:35:17roolkuSlasheri: yes, for storing pointers to disk
19:35:25*LambdaCalculus37 clones preglow into #rockbox-community
19:35:31roolkuSlasheri: I was thinking ahead on how to integrate it into the tree
19:35:34Slasheriroolku: storing dircache pointers in ram :)
19:35:37n1spreglow: i will look into it now, didn't know about it not working
19:35:50preglown1s: excellence
19:36:02rasherpreglow: Which is why I'm taunting it. For all its supposed unixness, it sure breaks a lot when you try to treat it like one
19:36:21Slasheriroolku: the idea with tag_filename is that the actual pointer to the index in the filename "table" is replaced with a dircache pointer from the ram copy of the db
19:36:27*amiconn should probably implement his idea to always compile the test_* plugins, just not zip them up by default
19:36:48preglowrasher: this seems to be a gcc build system bug, tho
19:36:57Slasherithat way the path can be fetched directly from dircache without ever touching the actual filename table after db loader is finished
19:37:09preglowthe last time i tried, it was autotools that completely messed up
19:37:12preglowhow i hate autotools
19:37:41roolkuSlasheri: I think I got the idea - I'll have a look how it is done for the filename
19:37:54Slasheriroolku: great :)
19:38:59rasheramiconn: that should be quite simple.. somewhere around line 300 in
19:39:45rasheramiconn: err, no, not there..
19:40:00amiconnrasher: It would probably be best to invent a new category for them, so they are added to rocks/apps, but only when gets an extra parameter from 'make'
19:40:06linuxstbWhat about just including them in the builds? e.g. accessible via the debug menu?
19:40:20linuxstb(putting them in something like rocks/debug/)
19:40:59amiconnNot all of them are debug stuff, and we should keep the downloadable packages small, imo
19:41:56Nico_Proolku: I just tested the patch... what would be ideal is to have a plain list of album covers
19:42:15amiconnHaving them in the normal locations, but only when making "devzip" or sth wouldn't add anything to the rockbox code, and also not to the download packages
19:42:33roolkuNico_P: not sure I understand - how is what you get not plain?
19:43:09Nico_PI only looked in the DB menus and all I see is "cover by..."
19:44:12roolkuNico_P: well, these are some example you could get all heavy metal covers if you do cover by genre for example
19:44:17 Join Horschti [0] (
19:44:28roolkuNico_P: or all covers by a specific band
19:44:57*linuxstb looks at the bmp vitja committed for the iaudio7 sim ;)
19:45:01 Quit Horscht (Nick collision from services.)
19:45:08Nico_Proolku: what I get is a list of tracks
19:45:51*roolku needs to patch and compile
19:46:58roolkuNico_P: first you select (say) a genre, then you get a lists of cover image filenames and if you select one you get a list of tracks for that cover (but filtered by genre!)
19:47:13roolkuNico_P: that is how it is supposed to work though
19:47:31 Quit Nibbl (Read error: 60 (Operation timed out))
19:47:32Nico_PI don't see the list of covers
19:47:42Nico_Pmaybe I screwed up my sync
19:47:45roolkuNico_P: instead of selecting a cover image path, you are supposed to select an cover image though
19:48:15bertriklinuxstb, what would be a good place to put the as3525 codec read/write code? I'm thinking of firmware/target/arm/as3525 (I think it's generic for all as3525 chips)
19:48:33bertrikand if the bootloader doesn't need it, would I have to take special precautions?
19:48:44linuxstbbertrik: Where is it in the e200 port?
19:48:50bertrik(like an #ifdef in some SOURCES file)
19:49:24linuxstbbertrik: No, you shouldn't need to take any precautions - if the code isn't called, the linker will discard it.
19:49:24 Quit J-23 (Remote closed the connection)
19:49:40linuxstbOr rather, if nothing in the .c file is used...
19:50:06bertrikthe e200 port uses the generic i2c-pp.c (or is it pp-i2c.c)
19:50:33linuxstbThen i2c-as3525.c ?
19:50:43linuxstb(in firmware/target/arm/as3525)
19:51:15bertrikthen the various parts that need to access the codec, use i2c calls. For as3525 I think we can abstract it as as3525_codec_read/write calls as the i2c bus it's on is basically an internal i2c bus
19:53:16roolkuNico_P: I'll try to sync the patch later - looks like another tag has been introduced since...need to go home and eat something first
19:55:20Nico_Proolku: great :) I'm really curious to see what kind of things we could do in pictureflow with better DB integration
19:59:12*Nico_P would love to see JPEG and resize in 3.1, but maybe that's a bit optimistic
19:59:42 Join karashata [0] (n=kimi@
20:08:49linuxstbNico_P: You've probably got about 6 weeks until feature freeze....
20:09:15Nico_Pthat's pretty tight...
20:09:47linuxstbOr maybe less, I forget when the 3.0 feature freeze happened...
20:10:16gevaertslinuxstb: that image looks like a great way to show which ports are not ready yet :)
20:10:36linuxstbgevaerts: Good idea... ;)
20:10:37pregloware we going to push releases even if nothing big happens?
20:11:42*Nico_P wonders what image gevaerts is referring to
20:12:07linuxstbiaudio7 sim background
20:12:46Nico_Phahaha, nice :)
20:13:03 Join DerDome [0] (
20:17:23 Join J-23 [0] (
20:17:29 Join mrpeters [0] (i=0cba500e@gateway/web/ajax/
20:17:52 Quit Thundercloud (Remote closed the connection)
20:18:01mrpetershey guys...quick one....recording line in on a ipod video 30gb...possible?
20:18:15mrpetersworks great on my iriver...but desperately seeking a second recorder to tape a show tonight :)
20:21:26n1slinuxstb: test_codec uses the regular codec buffer but a separate malloc buffer (just like we used to in the core) however to give a codec the rest of that buffer in test_codec would require to export some more functions to the plugin api, do you think it would be ok to just make that test_codec buffer 1MB and be done with it?
20:22:40linuxstbYes, I think so - assuming you mean to use CODEC_BUFFER_SIZE. test_codec is for benchmarking, so I don't think simulating memory usage is important.
20:23:45 Quit HBK (Read error: 110 (Connection timed out))
20:23:56n1syes, the define should be used of course
20:25:18 Quit BMBL (Read error: 113 (No route to host))
20:28:15gevaertsmrpeters: I'm not sure, but I think the ipod video has line in on the dock connector. You'd need an adapter of course
20:28:40 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
20:29:19BigBambiAnd I seem to remember it isn't really available on any bought adaptors and you have to make a custom one. Not sure about that though, I may well be wrong :)
20:30:29 Quit hannesd (Read error: 113 (No route to host))
20:30:29mrpetersthanks for the info :) making one doesnt scare me, just need some directions :)
20:30:39linuxstbYes, there's a line-in hidden in the dock connector, and I'm not aware of any commercial adapters that include it. But they can be bought if you google for them.
20:30:42 Quit {phoenix} (Remote closed the connection)
20:31:45 Join Nibbl [0] (
20:36:04 Join elinenbe [0] (n=elinenbe@
20:38:16mrpeterslinuxstb....any idea what to search for? getting no results for ipod video line-in dock adapter
20:38:55linuxstbremove the "video" - it's generic to all (or rather, most) ipods
20:39:17linuxstbProbably just "ipod line-in"
20:41:01 Quit stoffel_ ("leaving")
20:46:50mrpetershmm...wonder if i could make one of these in a matter of an hour or two haha
20:48:35 Quit J-23 (Remote closed the connection)
20:48:53linuxstbmrpeters: Probably - see for the dock pin description.
20:49:26linuxstbmrpeters: Here's a pre-made adapter -
20:50:21n1shmm, if anyone's really interested in finding out how much a codec mallocs that malloc buffer in test_codec could be filled with 0xdeadbeef or something and checked after the codec exits...
20:51:50 Quit mcuelenaere (Read error: 110 (Connection timed out))
20:53:04 Quit BigBambi ("Please insert girder")
20:53:04linuxstbn1s: Could be useful, although part of the problem is with very big files - which test_codec can't handle
20:53:12 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
20:53:44n1slinuxstb: true
20:58:51 Join einhirn [0] (
20:59:25 Quit BigBambi (Read error: 60 (Operation timed out))
20:59:54 Quit karashata (Read error: 104 (Connection reset by peer))
20:59:55 Join karashata [0] (n=kimi@
21:00:10 Part mrpeters
21:02:28 Quit nuonguy ("This computer has gone to sleep")
21:02:47 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:03:29 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:03:34***Saving seen data "./dancer.seen"
21:04:04amiconnlinuxstb: That could be a debug option, using logf()
21:04:49 Join n17ikh|Lappy [0] (
21:05:42 Join sarixe [0] (
21:07:01 Join Boopop [0] (n=892c01c8@gateway/web/cgi-irc/
21:07:18BoopopHi all, is there a way to get my rock'd iPod to play a new song every minute? (drinking game!)
21:08:15roolkuNico_P: I have synched the patch - what you have seen was most likely because you didn't have any .100x100 covers (and hence everything ended up as untagged) - it is now only searching for covers without size specification. This needs to be changed so it matches what picture flow can display
21:08:37roolkuNico_P: I will now look at what Slasheri suggested
21:09:09bluebrotherBoopop: make every song exactly one minute long
21:09:39Boopopbluebrothe : Well that's the long way to go about it, but I was hoping there might be a plugin to do it
21:09:44Boopopthanks all the same
21:09:49Boopopbluebrother* even
21:09:52Nico_Proolku: I'll check with the synched patch
21:09:53 Quit gevaerts (Nick collision from services.)
21:10:00bluebrotherno, there isn't an "easy" way.
21:10:03 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
21:10:17bluebrotherthis is also kinda esoteric. But you could make your own plugin if you like to
21:10:26BoopopOk, thanks anyway
21:10:49elinenbewhat's the status on the playstation emulator for rockbox?
21:10:57BoopopWell I'm about 3 weeks into a Computer Science degree, so making such a plugin is a bit beyond my grasp
21:11:01 Join MU{lappy} [0] (
21:11:13linuxstbBoopop: Rockbox is a fun way to learn programming...
21:11:22Boopopunless by some miracle rockbox runs on java and there's some magical line of code to do what I want, and I doubt that's the case
21:11:44BoopopWhat language do plugins use?
21:12:00domonokyand plugins are easy, but i dont know if all needed playback options are already in the plugin_api
21:12:46bluebrotherI'm pretty sure they are −− several plugins can skip forward / backward / pause playback
21:12:46domonokybut just controlling a already playing playlist should be pretty easy...
21:13:52domonokybluebrother: i think functions to create new playlists are missing, but controlling a already playing one, should be no problem..
21:15:02bluebrotherdomonoky: well, as he only wants to play a new song each minute controlling the playlist should be enough, right? ;-)
21:15:26 Quit reacocard ("bye")
21:16:13domonokyBoopops plugin could in the easiest solution just be a " while(1) { rb->audio_next(); rb->sleep(HZ*60); }" inserted into the helloworld plugin :-)
21:16:28 Quit Xerion (Read error: 104 (Connection reset by peer))
21:16:54 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
21:19:53BoopopIt's ok everyone, the website everyone was gonna use is back up
21:20:06bluebrotherdomonoky: yielding between should be necessary though
21:20:06BoopopThanks all the same, I think I'm doing C next year :P
21:20:10 Quit Boopop ("CGI:IRC")
21:27:15 Quit sarixe (Remote closed the connection)
21:27:43 Join sarixe [0] (
21:29:26 Quit mf0102 ("Ex-Chat")
21:35:00 Quit sarixe (Remote closed the connection)
21:35:44 Join sarixe [0] (
21:36:23 Quit Nibbl (Read error: 113 (No route to host))
21:40:29 Join meven [0] (
21:44:04 Join Seed [0] (
21:44:06Nico_Proolku: yep, it works as expected now
21:44:27Nico_PI might try to tinker a bit with PF
21:44:40roolkuNico_P: :)
21:44:55Nico_Proolku: does this also mean that each track has a cover associated with it?
21:45:06Nico_PI guess it does
21:45:24roolkuNico_P: if there isn't one then it is "untagged"
21:46:47Nico_Pyou mean the cover entry for the track will state "untagged"?
21:46:54amiconnNico_P: I would prefer if pictureflow would at least give meaningful messages
21:47:03 Join petur [50] (n=petur@rockbox/developer/petur)
21:47:10amiconnRight now it fails with a very confusing message if there are no images at all
21:47:13Nico_Pamiconn: pictureflow needs a lot of improvements
21:49:55 Quit Gareth (Remote closed the connection)
21:50:29Nico_Proolku: I added an "album" choice in "cover by..." and it seems to work just fine. this is exciting :)
21:50:51LambdaCalculus37I had opened a bug report about PF not refreshing its cache properly when you update... see FS #9135.
21:55:01roolkuNico_P: yes, databases are nice ... if this gets in, you could probably speed up the search for album art on track loading by just looking into the DB
21:55:27Nico_Pindeed. it should be *much* faster
21:56:07Nico_Proolku: especially with the dircache change Slasheri suggested
21:56:55 Join Thedjatclubrock [0] (
21:57:22ThedjatclubrockDoes RockBox read the iTunes dirs on the iPod
21:57:35LambdaCalculus37Through the database function, yes.
21:57:56roolkuNico_P: I am struggling a little - it needs changing in quite a few places - lets hope I find them all :)
21:59:23 Join BigBambi_ [0] (n=Alex@rockbox/staff/BigBambi)
21:59:43ThedjatclubrockWill RockBox fail with an iPod Speaker Dock, and can I dual-boot.
21:59:54 Quit LambdaCalculus37 (" ajax IRC Client")
22:00:28BigBambi_Depends on the doc, it might work with a patch, and yes
22:00:31scorche|shThedjatclubrock: depends on what you mean by "fail", and yes you can (this is covered in the FAQ and manual)
22:01:20 Join massiveH [0] (
22:02:17ThedjatclubrockBut the is the warranty violated :P
22:02:27linuxstbYou still have a warranty?
22:03:00ThedjatclubrockNot really.
22:03:08ThedjatclubrockBesides, the iPod is dying.
22:03:15 Join Thundercloud [0] (
22:03:36scorche|shwhat about them?...they are just another audio file..
22:03:58ThedjatclubrockIs there such a menu item. Sync like iTunes?
22:04:02linuxstbBut no, Rockbox doesn't have any features specific to podcasts.
22:07:09 Quit pixelma2 ("-")
22:07:22 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
22:07:47ThedjatclubrockSo, they do not have their own menu item?
22:08:14 Join Hillshum [0] (
22:08:17scorche|shno, but you could make one by editing the code or by using the database
22:08:31linuxstbOr by putting them in their own folder called "Podcasts"
22:08:40amiconnOr by just putting them into their own folder...
22:08:45*amiconn too sloow
22:08:47*linuxstb won!
22:09:58ThedjatclubrockAlright, I'll consider it.
22:10:32linuxstbThedjatclubrock: The only thing you will lose is time - if you don't like Rockbox, it's easily uninstalled, and you'll get a 50% refund.
22:10:42*BigBambi_ doesn't get what is so special about podcasts
22:10:55 Quit BigBambi (Read error: 113 (No route to host))
22:11:10scorche|shlinuxstb: i think we should up that to 75% refund now
22:11:39ThedjatclubrockWell, I need to have it fixed, and then I will RB it.
22:12:24ThedjatclubrockI hope iPod G5 battery is ok
22:13:56ThedjatclubrockNo HFS+ support :/
22:14:13scorche|shwe are aware..
22:17:37ThedjatclubrockI need to boot the iPod OS to charge it?
22:18:57scorche|shno...have you read through our fine manual?..if not, i suggest you do
22:19:23 Quit BigBambi_ (Remote closed the connection)
22:20:34dionoeascorche|sh: is a tower of rockbox planned at the gsoc mentor summit?
22:20:41dionoeaIs anyone else from rockbox going btw?
22:20:52ThedjatclubrockAww, no nano g2 support
22:21:29scorche|shdionoea: markun, but i dont know how many targets he is bringing and i wont be bringing my whole collection
22:21:29Bagderdionoea: scorche and markun
22:21:48dionoeaI can bring 3 if you want :) (archos recorder, ipod video and sansa e200)
22:21:54Bagderthat devcon tower is hard to beat... ;-)
22:22:03scorche|shdionoea: sure ;)
22:22:06dionoeacustoms officials might wonder why i have so many daps though :)
22:22:17dionoeaBagder: indeed :p
22:22:33scorche|shdionoea: perhaps i will bring more targets then if we might get a decent tower going
22:22:51domonokyBagder: just wait for next devcon. we will build the twin towers of rockbox there :-)
22:22:59 Join AJCantos [0] (
22:23:03scorche|shdomonoky: with flying buttresses!
22:23:08Bagderyay, with a bridge between them!
22:23:15dionoeaand planes? (sorry)
22:23:23scorche|shoh dear..
22:23:38scorche|shdionoea: were you planning on getting past airport security? ;)
22:23:50dionoeahum ... I have to remember to behave :)
22:24:01*scorche|sh uploads this chatlog to the report
22:24:03dionoeaI'm leaving home at 5:30 AM tomorrow to get my plane :(
22:24:18dionoeado you have 7hours left to contact the customs :)
22:24:20 Quit pierre- (Read error: 110 (Connection timed out))
22:28:58 Quit bmbl ("Woah!")
22:32:41 Quit miepchen^schlaf ()
22:33:00 Join AndyI [0] (i=AndyI@
22:34:41ThedjatclubrockiPodLinux vs Rockbox
22:35:02scorche|share you actually asking anything, or just saying random things?
22:36:08Bagderand the question is...?
22:36:08scorche|shwhat are you asking?
22:36:11linuxstbNo, that's a fixture.
22:36:27ThedjatclubrockBut anyway, if I format my iPod FAT, and I use it on the mac, will the DB be able to be converted with the windows tool?
22:36:56linuxstbWhat DB? What windows tool? converted to/from what?
22:37:45linuxstbThat tool is out of date, and isn't needed anyway - Rockbox will build its own database based on the tags in your audio files.
22:37:55 Quit bertrik ("Leaving")
22:38:15ThedjatclubrockTags in the iTunes "Get Info" dialog.
22:38:34linuxstbTag in the files. We don't know what itunes displays.
22:39:27 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
22:40:36 Join mirak [0] (
22:40:37ThedjatclubrockWhere would I put the files if I didn't use iTunes.
22:41:09ThedjatclubrockAlso, can RB sort by Album/Artist etc...
22:41:51domonokyThedjatclubrock: please take a look at the manual. its all in there
22:42:02BigBambiPut them wherever you want, and yes - please please please rtfm
22:42:13ThedjatclubrockSorry guys.
22:42:17ThedjatclubrockI'll bbl
22:48:07 Quit neddy (Read error: 104 (Connection reset by peer))
22:57:51pixelmalinuxstb: maybe Riyonuk's sim really crashed due to the 0 pixels wide progress bar. Need to try...
22:59:31 Quit massiveH ("Leaving")
23:00:26linuxstbpixelma: Yes, I looked at that code last night, and think it would lead to a divide-by-zero
23:01:12 Join kb__ [0] (
23:02:32 Join krazykit [0] (
23:03:36***Saving seen data "./dancer.seen"
23:04:20 Quit kb__ (Client Quit)
23:04:45 Join L29Ah [0] (n=L29Ah@
23:07:13 Quit domonoky (Read error: 104 (Connection reset by peer))
23:09:55 Quit jgarvey ("Leaving")
23:15:19 Quit karashata ("I go, only to return again some time...")
23:16:35n1she did say he got a div by 0 exception on the player...
23:17:49n1samiconn, preglow or anyone familiar with coldfire asm could you remind me why move.l (%a2,%a0.l*4),76(%sp) is illegal on our coldfires?
23:18:01 Quit elinenbe (" HydraIRC -> <- The future of IRC")
23:18:41 Join neddy [0] (n=john@nat/sun/x-49495d8d73e9f956)
23:18:47 Join fml [0] (n=4fd3c122@gateway/web/cgi-irc/
23:19:07linuxstbn1s: (and pixelma) - I'm about to commit a fix for that - the wps parser should reject a 0 width (or height)
23:19:29pixelmaI get a freeze when trying to start playback with a WPS that contains a progress bar with a wide of 0 - freezes at the "Loading..." splash
23:19:47linuxstbIn the sim or on target? In Linux, the sim crashes with a floating point exception
23:19:52n1slinuxstb: that sounds like the Right Thing (tm) indeed
23:20:02pixelma*width too, that's in a c200 sim
23:20:51 Quit roolku ()
23:21:11fmlXavierGr: I am here but not for very long. Would you like to discuss FS #9455?
23:21:26*linuxstb also takes the opportunity to add some whitespace, and reminds JdGordon|zzz that it's free ;)
23:21:50 Join LambdaCalculus37 [0] (
23:22:37pixelmaonce it even started playback for a few seconds, then stopped with a "sdl_audio_callback: No Data." but I can't reproduces, usually it just freezes without any debug info
23:24:12*linuxstb isn't sure what width/height actually mean when there's a bitmap...
23:24:35linuxstbSetting a 0 height doesn't seem to make any difference (in cabbiev2)
23:26:03*pixelma tried with a non-bitmapped progress bar (own WPS)
23:28:56 Quit bughunter2 ("bye")
23:29:55pixelmaI also discovered that there is a 70 line (about) limit for .wps file length (needed to "optimise" some pure comment lines)
23:31:45fmlHe-he, I seem to have found a bug in the settings storage code. I think that if the name of, say, wps file has a '.wps' in the middle (e.g. my.wps.file.wps) then it will be stored as just 'my' in the cfg file. I'll try it out.
23:31:56 Join Slack [0] (
23:32:16amiconnn1s: Umm, because it is?
23:32:54amiconnCFPRM.pdf is a bit vague on this, but the MCF5249 user manual (instruction timing tables) clearly show that it's not allowed
23:36:25n1sah, so you can't use a scale factor on one operand and dicplacement on the other _at the same time_ thanks
23:37:31n1sI am starting to consider diving into gcc and trying to fix a bug that causes it to spit out that instruction when compiling for coldfire...
23:37:51fmlHrm, no, it's stored correctly. Or not? It's been written as wps: /.rockbox/wps/my.wps.file.wps Shouldn't the '.wps' suffix be cut off? Ok, it's not what I expected but still.
23:38:12amiconnn1s: Eh? When/where does it do that?
23:38:24n1sthis only happens on gcc4.3 >= afaik but it is triggered by instruction scheduling so it's probably latent in earlier versions
23:38:53 Quit Schmogel (Read error: 104 (Connection reset by peer))
23:39:03linuxstbpixelma: Are you sure it's a line limit, and not a limit on the total size (in bytes) of the .wps file?
23:39:17*amiconn has given up on hoping for improvements in newer gcc versions
23:41:15n1samiconn: I'm not that jaded yet, so i still keep on hoping but my hope got a bit lower today, as building a m68k crosscompiler failed completely with 4.3.2, 4.3.1 did build if i used sed extensively to correct paths after configuring...
23:42:18amiconnThey promise improvements with every new release.... what you get instead are mainly new useless warnings and new quirks to work around
23:42:18pixelmalinuxstb: not completely
23:42:22 Quit neddy (Read error: 54 (Connection reset by peer))
23:42:41linuxstbpixelma: The line limit seems to be ((LCD_HEIGHT/5+1) * 2)
23:42:42amiconn4.0.x was the last real improvement I've seen (wrt size optimisation - probably the tree-ssa stuff)
23:43:08linuxstbpixelma: But that shouldn't include comments.
23:44:03amiconnBut even that one intriduced a quirk I had to work around for archos
23:44:36 Quit petur ("Zzzzz")
23:45:07pixelmalinuxstb: basing this off of LCD_HEIGHT only seems unfair to me - I didn't change anything in the real WPS code, just cutting some comments and whitespace
23:46:21linuxstbpixelma: IIUC, that line limit doesn''t include comments - just successfully parsed lines
23:46:21pixelmaaha, commenting out whitespace lines also helped, ok
23:46:51 Join Zarggg [0] (
23:46:55n1samiconn: I agree that new versions of gcc often do not bring anything but problems, I've done a bit of testing for coldfire but from what i found with limited performance tests, 4.3 is the best 4.x so far (coldfire)
23:47:23*linuxstb wonders if the builds are just slow, or broken...
23:48:38 Quit Nico_P (Remote closed the connection)
23:49:13amiconnn1s: I'd just go 4.0.3 for colfire for uniformity (maybe go 4.0.4 instead for arm, cf and sh1 because it's the latest 4.0.x release).
23:49:39amiconnI also wonder whether the SH1 bug is fixed in later versions. I sincerely doubt it is...
23:49:56ThedjatclubrockCan I somehow export my iTunes playlists to rockbox?
23:51:25n1samiconn: fixes for less commonly used platforms seems to be few and far between unfortunately :/
23:51:28pixelmalinuxstb: still using LCD_HEIGHT as a base for this calculation will give different limits for the same sized screens - e.g. on the e200 the limit will be higher than on the H300/Ipod Color with the same sized screen. Now with viewports the argument that you could fit more lines (with the same font height) on the screen isn't so strong anymore...
23:52:19n1s(i reported that bug i mentioned in febuary and got a comment by one of the maintainers basically stating where the bug is and then nothing)
23:53:08amiconnn1s: Problem is that I don't understand gcc code well enough to provide a proper fix. My simple patch is working reliably though. And I really don't want to contact those guys again, given earlier experiences...
23:54:33 Quit ender` (" I think I remember an episode of MacGyver where he overthrew a violent dictator with a rubber band, 2 bottle caps, and some")
23:57:29ThedjatclubrockCan I somehow export my iTunes playlists to rockbox?
23:57:40n1sThedjatclubrock: no
23:58:23ThedjatclubrockAlso, does the RB db update after a USB connect. Does ut need to search everyting to uodate?
23:58:29linuxstbBagder: The build system seems stalled... Will it fix itself?

Previous day | Next day