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 2006-02-03

00:00:33 Quit Midgey34 (Read error: 110 (Connection timed out))
00:00:36Midgey31what will that code return? columns?
00:00:38t0masafaik you can't get the size of an array...
00:00:44 Nick Midgey31 is now known as midgey34 (
00:00:48t0masthat would return the size of the datatype afaik?
00:00:53ep0chcutoff = 0hz is interesting
00:00:59dpassen1You'll get the size of a pointer.
00:01:03t0masso for an int[100] it would just return 4
00:01:09t0masah, always 4
00:01:19 Quit midkay (Remote closed the connection)
00:01:45midgey34well darn
00:01:47preglowcutoff = 0hz should not be possible
00:02:14t0masmidgey34: is the array predefined?
00:02:20ep0chsounds like some peoples cars i know of
00:02:25t0masbecause then you shouldn't need to find out the size :)
00:03:14t0masand if it's not... can't you keep a counter of how many elements you have?
00:03:51midgey34I'm going to have to go the second route, it just adds more variables to the game context
00:04:07t0masgood night, and good luck :)
00:04:14 Quit t0mas ("have to get up early tomorrow")
00:05:02preglowsomethings wrong here
00:05:07 Quit bluey (Remote closed the connection)
00:05:39 Quit Siku ()
00:05:55preglowthe eq filters should not self-oscillate this often...
00:06:39Philip_0729how do you set the backdrop??
00:06:41ep0chany values in particular?
00:07:53preglowep0ch: they do so for everything i do now'
00:07:59preglowep0ch: i've messed up the patch in some way
00:08:09preglow'cause everything worked like expected an hour ago
00:09:13preglowthis will have to wait, i'm falling asleep
00:10:59 Join ts-x [0] (
00:11:06 Quit slimx (Remote closed the connection)
00:11:48MoosGood night preglow
00:12:19 Quit San||Study ()
00:12:36elinenbeLinusN: uh oh! I guess your X5 is working! ;)
00:12:50LinusN:-) nah
00:13:15peturrecording gain patch updated and ready:
00:14:52MoosLinusN: do you have already BDM connected in X5?
00:16:15Rob2222Would it be much work to make the RB text color selectable? Now we have backdrops...
00:16:47LinusNi haven't decided on a good approach for it yet
00:17:00JdGordonthe h300 optimized build has it
00:17:20LinusNthat patch is not a good way to do it imho
00:17:52JdGordonhow do u set the background image?
00:17:56LinusNmostly because it's hopeless to maintain the localization
00:18:00Rob2222LinusN: Is it so hard to implement?
00:18:16LinusNJdGordon: hold navi on the .bmp file
00:18:32LinusNRob2222: no, it isn't hard, it just has to be done
00:19:06Rob2222Has no one made a (good) patch for it, yet?
00:19:09LinusNi'd gladly accept an RGB slider setting patch
00:19:41Rob2222Hmm, ok.
00:20:14JdGordonLinusN: then what? or is it not in the latest builds yet?
00:20:24LinusN3 horizontal sliders on one screen, a display of the combined (hex) value, and a visual representation of the selected color
00:20:31peturtime to hibernate a bit
00:20:35 Quit petur ("here today, gone tomorrow")
00:20:55LinusNJdGordon: in the context meny, select "set as backdrop"
00:21:13JdGordonnop, only got the usuall items
00:21:14LinusNit's in the bleeding edge
00:21:39Rob2222LinusN: When Ive done my tests in university I will fresh up my C knowledge and maybe I can contribute anywhen. But atm my tests have first priority for the next month.
00:21:39JdGordondoes the "latest" link to bleeding edge or something else?
00:21:54LinusNwhat "latest" link?
00:22:00JdGordonunder the pic
00:22:19Rob2222thats the daily under the pic
00:22:22LinusNJdGordon: that's the daily build section
00:22:28Rob2222you must use the link more down
00:22:38LinusNgo to the bottom of the page
00:23:07*JdGordon is still asleep
00:23:16*LinusN should be
00:24:18 Part ep0ch ("Kopete 0.11.1 :")
00:25:39JdGordonhmm.. can u unset the background img?
00:25:47LinusNyes, in the lcd settings
00:26:11Philip_0729looks very cool Linus well done once more :D
00:26:32 Part Philip_0729
00:26:38JdGordon:) looks cool.. but the image i whipped up doesnt.. so ill wait for some1 with more graphic talent than me to make a nice 1
00:26:50 Quit NightCat ()
00:27:46LinusNtime to sleep, nite all
00:27:59 Part LinusN
00:28:29BHSPitMonkeyJdGordon: what image?
00:28:49JdGordoni took the joltamp background and hacked it up a bit to make it look like a border
00:28:55JdGordonneeds fiddling
00:31:35Paul_The_NerdWell, since the text always displays on the outer edge right now, a border doesn't work so well, 'eh?
00:31:52JdGordonno, i forgot that when i was making it :p
00:32:15Paul_The_NerdThere *was* some talk about being able to set a viewport in which the menus were displayed, but nobody's done anything with that talk.
00:34:42amiconnlinuxstb: Just an idea worth thinking about, imho:
00:35:00amiconnWould it be possible to require the backdrop image buffer to be line aligned?
00:35:17amiconnIf the answer is yes, there is probably a lot of potential for speedup
00:35:38amiconnNo hurry, I won't be able to do anything in that direction today
00:38:32Rob2222good night all
00:39:19 Join linuxstb_ [0] (
00:39:24amiconnBtw, background clearing doesn't need to use the clearpixel functions at all
00:39:40linuxstb_amiconn: Yes, that's no problem at all - the buffers are defined in apps/recorder/backdrop.c
00:39:58linuxstb_So you can align them any way you wish.
00:40:07amiconnBut later; now -> sleep
00:40:13BHSPitMonkeyPaul_The_Nerd: that's neat
00:40:19BHSPitMonkeyJdGordon: ^
00:40:48JdGordoni got rid of the left and top line and it looks better... im gonna keep playing
00:41:17 Join Cassandra [0] (
00:41:56CassandraOK - so how do I tell Rockbox to use a backdrop?
00:42:24 Join San [0] (
00:42:53Paul_The_NerdCassandra: Create a bitmap equal width and height to your screen. Then, in rockbox, long-hold on it to get the menu, and choose Set as Backdrop.
00:42:55linuxstb_Just browse to it, and select "Set as backdrop" from the onplay menu
00:43:14Paul_The_NerdAlso, there' the %X tag for use in WPS files.
00:43:23linuxstb_If you want Rockbox to remember it, it has to be in /.rockbox/backdrops/
00:43:25dpassen1do backdrops work on the H1x0?
00:43:40linuxstb_dpassen1: Not yet, but they are planned.
00:43:54CassandraOh yeah, baby!
00:43:59Cassandradpassen1, nope.
00:44:08CassandraThat is a thing of beauty!
00:46:59 Join San||Study [0] (
00:47:46 Quit CoasterMaster ()
00:49:05 Nick San||Study is now known as [San] (
00:54:55lostlogicHmm... the pcm_* functions currently in the codec API really want removing. codecs should have no direct access to pcm.
00:56:39 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
00:58:02 Quit linuxstb_ ("CGI:IRC")
00:59:04 Nick ashridah is now known as Lost-ash (
00:59:08linuxstblostlogic: I would say that if there are any functions in the codec API not being used, then delete them. AFAIK, there are no third-party codecs to worry about.
00:59:22 Quit ender` (" I think that's how Chicago got started. A bunch of people in New York said, 'Gee, I'm enjoying the crime and the poverty, bu)
00:59:36lostlogiclinuxstb: w00t, I'll be doin' that as part of this whole audio engine audit then... if I ever make it stop deadlocking.
01:00:46linuxstbThere's no reason not to do it before then. Other people have criticised the codec api as well for containing too much.
01:00:55 Quit Paul_The_Nerd (Read error: 104 (Connection reset by peer))
01:02:22lostlogichmm... *nod* I'll see what I can do... also should get the pcmbuf part of my work in regardless of the state of the playback part... better to do things in smaller bites.
01:02:40 Join optimistik [0] (
01:02:51linuxstbI agree with that.
01:02:56 Join Paul_The_Nerd [0] (
01:03:00 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
01:03:09optimistikcan someone tell me how to uninstall rockbox from my ipod?
01:03:48Paul_The_Nerdoptimistik: Did you save the bootpartition.bin you extracted when making the bootloader?
01:04:39 Quit San (Read error: 110 (Connection timed out))
01:05:00CassandraUse Apple's firmware restore utility.#
01:05:08Paul_The_NerdIf you use the same instruction you used to install rockboot.bin but instead replace it with bootpartition.bin, that should restore your firmware partition to the state it was in before. Then you can delete .rockbox and rockbox.ipod
01:05:12CassandraThat's the easiest way.
01:05:17Paul_The_NerdCassandra: That way also wipes the HD.
01:05:37CassandraPicky, picky.
01:05:42Cassandra(Oops, I'd forgotten that.)
01:05:53***Saving seen data "./dancer.seen"
01:06:19optimistikk thanx
01:06:22Paul_The_NerdI've used the restore utility a few times out of necessity, but dding or ipodpatching the bootpartition.bin is actually faster.
01:06:31Paul_The_Nerdoptimistik: Out of curiosity, why are you wanting to remove it?
01:07:45optimistikum i can't boot into the apple os cuz i got ipod linux on there too and i'm suposed to push menu to boot into the normal apple os but when i do that it boots linux
01:08:12Paul_The_NerdAre you using the Rockbox bootloader, or iPodLinux's loader?
01:08:26Paul_The_NerdMenu boots the Apple OS with ours, and Play boots Linux
01:08:57Paul_The_NerdI'm not sure how you boot the Apple OS with theirs though.
01:09:54optimistikdo u push menu at the screen or when it shows the apple logo?
01:10:08Paul_The_NerdThe Apple logo.
01:10:27Paul_The_NerdThe easiest way is to actually tap menu to turn it on, and then immediately press and hold it until you see the words "Loading Original Firmware"
01:10:50optimistikif i do that it boots linux and if i don't push anything it boots rockbox
01:10:54 Quit Moos ("Glory to Rockbox!")
01:11:04Paul_The_NerdAnd you're using the Rockbox Bootloader?
01:11:19optimistiki'm not sure how do i kno?
01:11:42Paul_The_NerdDid you download it from
01:11:52Paul_The_NerdAlso, did you have linux on your iPod before installing Rockbox?
01:12:22Paul_The_NerdAnd I'm guessing you didn't uninstall it first.
01:12:48Paul_The_NerdWhich means that you *probably* merged our bootloader with the iPodLinux bootloader.
01:13:52optimistikhmm i guess so lol and i'm guessing that was stupid
01:14:12Paul_The_NerdI wouldn't say stupid. Just, a case that had not been predicted.
01:14:21linuxstboptimistik: Did you already have ipodlinux installed before you installed Rockbox?
01:14:26linuxstbI think that's the problem - when you press MENU in the Rockbox bootloader, it will start the original firmware that was on your iPod when you installed the Rockbox bootloader. In your case, that original firmware contained the ipodlinux bootloader - so the Rockbox bootloader starts the Linux bootloader, which starts Linux......
01:14:32Paul_The_Nerdlinuxstb: He already said yes to that.
01:15:22 Join tvelocity [0] (
01:16:16linuxstbPaul_The_Nerd: I wrote that about 5 minutes ago :) Strange things are happening in my irc client....
01:16:56Paul_The_Nerdlinuxstb: Aaah. Man, did you go and get your computer possessed again?
01:16:59BHSPitLappywhat's gdb stub in the configure utility?
01:17:18linuxstboptimistik: If I was you, I would follow these instructions to restore the apple firmware:
01:17:53linuxstbWait - that will probably delete your Linux partition.....
01:18:14BHSPitLappyoptimistik: what OS?
01:18:39optimistikits ok if idelete the linux i don't use it that much just to play a video once in a while
01:19:06BHSPitLappyoptimistik: do you have windows?
01:19:49BHSPitLappythere's a program for windows that can extract the apple_os for you, from the apple updater.exe
01:20:00BHSPitLappyso you don't have to restore the ipod or anything
01:22:40optimistikafter i get it off is there a way for me to have linux and rockbox on there?
01:23:31BHSPitLappythat was my point
01:23:45BHSPitLappyto keep you from having to overwrite your linux partition
01:24:09BHSPitLappydownload that app, extract your ipod's firmware, then do the ipod_fw stuff to it
01:25:25optimistikbut then how do i get it to use rockbox's bootloader after instead of linux's so i can boot all three os's?
01:27:44BHSPitMonkeythat's a nice photo
01:28:06JdGordondamn... i never knew ipod's looked that good :D
01:28:33JdGordonnot as nice as my h300 tho :p
01:28:37BHSPitMonkeyoptimistik: the same way as before, but instead of using your ipod's backup, you use what you get from ipodwizard
01:29:07BHSPitMonkeyanyone tried a color wps on a 5G?
01:29:17Cassandrachristi/misc/iPod-Rockbox2.JPG"> may be a slightly better representation (although it's a little out of focus.)
01:29:22midgey34Cassandra: I love that cd
01:29:38midgey34the muse one
01:29:54JdGordonahaha.. only a little out of focus :D
01:30:27 Quit matsl ("Leaving")
01:30:29linuxstboptimistik: You can follow the instructions for the Rockbox bootloader, but instead of using the apple_os.bin that you extract from bootpartition.bin, use the apple_os.bin that ipodwizard extracts for you from the Apple updater.
01:31:12linuxstboptimistik: And then you simply need to copy a linux kernel (called linux.bin) onto your FAT32 partition, and holding PLAY whilst booting will start that kernel.
01:31:12BHSPitMonkeylinuxstb: he still needs to ipod_fw -o apple_os -e 0 firmware_from_ipodwizard
01:31:36linuxstbBHSPitMonkey: Yes - that's part of the instructions for installing the Rockbox bootloader.
01:31:59BHSPitMonkeyjust don't want him to think the wizard one is already stripped
01:32:15linuxstbWhat do you mean?
01:32:42linuxstbAh, sorry. So ipodwizard gives you a firmware partition image, not the apple_os.bin ?
01:32:51BHSPitMonkeywell, there's your firmware partition backup BEFORE you strip out the retailos, and then there's the one AFTER you do the -e 0
01:33:06BHSPitMonkeysort of
01:33:07JdGordonlinuxstb: can i assume that all the members in a sturct screen * will always be valid? (i mean the data members.. not funcitons)
01:33:09BHSPitMonkeyI think so, yeah
01:33:21linuxstbJdGordon: I've no idea about that part of Rockbox.
01:33:30CassandraIt's damn hard to get a decent photo of a colour LCD. Particularly if you've only got a compact. (It's a good compact, but still.)
01:33:32JdGordonany1 here that does?
01:33:48BHSPitMonkeyCassandra: what about a scanner?
01:34:03JdGordonamiconn: do u know?
01:34:39 Join lostlogi1 [0] (
01:35:15linuxstbJdGordon: He said goodnight a while ago...
01:35:52linuxstbYou should look at the code and find where the screen structures are initialised.
01:36:30linuxstbi.e. screen_init in screen_access.c
01:36:30 Nick Lost-ash is now known as ashridah (
01:37:36 Quit optimistik ()
01:38:51BHSPitMonkeyI just registered in the wiki, it says "A confirmation e-mail has been sent to"
01:38:59 Quit lostlogic (Nick collision from services.)
01:39:01BHSPitMonkeywhy did it do that to the suffix?
01:39:02 Nick lostlogi1 is now known as lostlogic (
01:39:16linuxstbDid you get the email?
01:39:40CassandraHmm - scan's not bad, but it kind of washes out the display
01:39:56BHSPitMonkeyyeah, but I've never seen that done to my address
01:40:24BHSPitMonkeyCassandra: try when your nano's lost its luster ;)
01:40:45linuxstbCassandra: I assume that boxes now uses a backdrop image?
01:41:21CassandraMine never had a lustre. It's an object trouvee
01:41:54Cassandralinuxstb, nope. I'll add backdrop setting support to the themes engine at some point though.
01:42:06Paul_The_NerdCassandra: You can use a %X tag in a WPS
01:42:08Paul_The_NerdCapital X
01:42:32CassandraPaul_The_Nerd, yeah, but if I do that, people can't substitute their own backdrop.
01:42:47CassandraAnd boxes is backdrop agnostic.
01:43:10Paul_The_NerdWell, I think inside boxes, with the screen pretty full of bitmappery, and pre-toned shadows, substituting backdrops may not be so likely
01:43:19 Quit darkless (Read error: 104 (Connection reset by peer))
01:43:24*Paul_The_Nerd shrugs.
01:43:43 Join darkless [0] (
01:44:10Paul_The_NerdI humbly disagree on the "backdrop agnostic" statement. :)
01:44:22CassandraPossibly. The %X tag is intended for WPSes that *require* a particular backdrop though.
01:44:29Paul_The_NerdI suppose that's true
01:44:41CassandraYou could just as easily use a green gradient with boxes, if you wanted.
01:45:27Paul_The_NerdBut, it looks horrible with say... the image of Thomas Edison I've put on the right side of my screen, for example. He watches over my menus.
01:46:11CassandraWell Paul, it's possible that we could have a seperate WPS backdrop setting, but that seems a little complex.
01:46:31lostlogichmmmmmm is there some way that the codec is being overwritten while it's asleep and that is why this thing crashes?
01:46:33CassandraIf it bothers you you can always do your own boxes with the %X tag in.
01:46:34 Quit mirak (Remote closed the connection)
01:48:12Paul_The_NerdI'm trying to think of it from the "Joe Average" perspective, who sets some cool looking menu background, then runs boxes and says "eww, there must be a bug, my backdrop is getting all over my WPS." I've already got a personalized WPS, so it's not a real concern to me, it just feels like the "demonstration" WPS shouldn't have what to users will be unexpected behaviour.
01:49:15 Quit youngcereal (Read error: 110 (Connection timed out))
01:50:05*Paul_The_Nerd will shut up now, before he gets too annoying. :)
01:51:01CassandraI see your point, but when we have foreground and background colour setting the WPS will alter as well. I think it's better to be consistent.
01:53:02 Join youngcereal [0] (
01:53:56 Quit Jolt ("CGI:IRC (EOF)")
01:54:22Paul_The_NerdI imagined you'd have tags in the WPS to override those as well, when the time comes.
01:56:43CassandraWell, early days. We'll see how it shakes out.
01:57:03BHSPitMonkeylinuxstb: ping
01:57:06 Join CoasterMaster [0] (
01:57:55linuxstbBHSPitMonkey: ?
01:58:15BHSPitMonkeyI made a small addition to the Ipodinstallation wiki page, wanted to see if you approve
01:58:29BHSPitMonkey(first wiki edit, feel awkward)
01:58:40BHSPitMonkeysection 1c
01:59:03 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
02:02:21linuxstbBHSPitMonkey: I think that page needs to be kept as simple as possible - I'm not sure if that will help or confuse people trying to follow those instructions.
02:02:52linuxstbMaybe it would be better if that was in the FAQ - a question about how to have both ipodlinux and Rockbox installed, and the installation instructions could link to the FAQ.
02:02:57BHSPitLappywell, the note specifies to whom it's directed
02:04:45midgey34is alac up to date with the original sources?
02:05:01linuxstbBHSPitLappy: I just think that in general, unneccessary choices are a bad thing - those instructions are complicated enough already.
02:05:11midgey34David Hammerton's Apple Lossless (ALAC) decoder ver 0.1.1 was released Nov 26
02:05:12linuxstbmidgey34: I wasn't aware of any updated releases. Have there been?
02:05:16BHSPitLappyI'll trust your judgement then
02:05:25BHSPitLappydo you want me to go in and remove it then?
02:06:19midgey34he mentions performance improvements on his site but I haven't taken much of a look at the source
02:06:45linuxstbBHSPitLappy: You could move it to the FAQ - you have space there to write as much as you want.
02:07:18 Quit hardeep ("My damn controlling terminal disappeared!")
02:07:46linuxstbmidgey34: Thanks for that link, I'll have a look. It comes at a good time - ALAC isn't realtime on the ipod for some reason.
02:08:12linuxstbOur source has changed a little from the original release - so it may be a little work to update it.
02:08:15midgey34that's why I decided to look into it
02:08:32midgey34alright, hopefully the changes are helpful
02:09:58CoasterMasterIs it just me, or does the shuffle function (at least on the iPod Nano) not really that random.....?
02:10:28Paul_The_NerdCoasterMaster: What exactly do you mean?
02:11:46linuxstbmidgey34: I've just looked at the differences, and they seem to me mostly in the metadata parser. Only one small function has actually been optimised, and there seems to be one bugfix.
02:11:55 Join topbloke [0] (
02:13:30linuxstbmidgey34: So I'm not sure why his release notes say "vastly improve improvements".
02:13:53linuxstbI mean "vastly improve performance"
02:19:12midgey34stream.c and stream.h have some changes, nothing that would affect us though
02:19:47midgey34I'm not seeing this vast improvement either
02:21:44 Join YouCeyE [0] (
02:24:05 Quit tvelocity ("Αναχωρεί")
02:24:23linuxstbThere is some optimisation in the deinterlace_16() function - but I've already changed that function. I can see some more simplifications now though, which I can test.
02:27:56midgey34do the changes to count_leading_zeros() affect us?
02:28:20midgey34in alac.c
02:28:45linuxstbMaybe - I haven't tried them yet. We need an ARM assembler version of that function.
02:28:49*linuxstb wakes preglow up
02:31:38JdGordonok, im almost done a colour picker screen... how the heck do i test it?
02:32:34linuxstbAdd it to a menu - any menu :)
02:33:00linuxstbOr make a plugin out of it.
02:33:58linuxstbmidgey34: Yes, the count_leading_zeros() optimisation was the big one. ALAC's now fine on the ipod.
02:35:21linuxstbSorry, not quite fine. It still struggles if data is being read from disk at the same time. But it's close.
02:37:07midgey34well good
02:37:35*midgey34 prays that the FFmpeg people are working on AAC
02:40:03 Join Yoo`Hoo [0] (
02:40:42 Part Yoo`Hoo
02:43:50linuxstbIf we can get asm versions of the count_leading_zeros() function for ARM and Coldfire, that should give a nice boost. But I'll commit the new C version from alac 0.1.1 now.
02:46:11 Join mafj2 [0] (n=mafj2@tor/session/x-4b186a08c10b4bac)
02:48:35 Join midkay [0] (
02:51:10 Quit paugh ("Leaving")
02:51:16CoasterMasterPaul_the_nerd: sorry it took so long to get back to you....but I was using Rockbox (the 02012006 daily) on my nano and the suffle seemed to do the same songs in the same order (I heard Hot Hot Heat's You Owe Me an IOU followed by Michael Franti's cover of What I Got a lot to day)
02:51:34 Join Midgey31 [0] (
02:51:44Paul_The_NerdCoasterMaster: As in, it ran through the playlist, started again, and the order did not change?
02:51:58Paul_The_NerdOr did you relaunch the playlist, and it actually said "shuffling" and then the order was still the same?
02:52:29CoasterMasterI relaunched the playlist and got the shuffling message
02:53:19Paul_The_NerdAnd this happens consistently that songs are often paired together? Statistically speaking, the odds are the same for X to be paired with Y a second time as they were the first time, so it could happen.
02:53:27CoasterMasterbut it wasn't a was a folder
02:53:50Paul_The_NerdWell, folders become playlists when you launch a song from them.
02:53:55Paul_The_NerdHow many songs are in the folder?
02:54:19CoasterMasterNot too many, which means it's probably a coincidence: 15
02:54:34Paul_The_NerdYeah, it's very possibly coincidence.
02:55:12CoasterMasterYeah, I'm playing with it some more and it seems to be shuffling fine
02:55:16linuxstbMidgey31: I've just committed that alac optimisation - thanks again for pointing it out.
02:55:25 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
02:55:28Paul_The_NerdOkay, I'm glad then. But thanks for the concern.
02:55:40CoasterMasterAlways trying to help out a good project :).
02:56:00Midgey31linuxstb: have you taken a look at the demux.c changes?
02:57:54linuxstbNot yet - I may look at them sometime, but Rockbox needs a better m4a/mp4 parser anyway - for the AAC codec.
03:00:30linuxstbAnyway, time for bed. Goodnight.
03:03:17BHSPitMonkeylinuxstb: is there a way to revert the wiki page, other than re-editing it
03:05:54***Saving seen data "./dancer.seen"
03:06:37 Part topbloke
03:07:16 Quit CoasterMaster ()
03:08:49 Quit midgey34 (Read error: 110 (Connection timed out))
03:10:07linuxstbBHSPitMonkey: Not as far as I know - just re-edit it.
03:10:18 Nick Midgey31 is now known as midgey34 (
03:16:47jaebirdI noticed slimx patch for 3g on the patch tracker from Jan 22 can the lcd portion be added even before the ata stuff is working for us 4g 2bpp users?
03:18:09linuxstbjaebird: That's a possibility. But there are no developers with a 4g who can test such a build...
03:18:58jaebirdlinuxstb: I can test the is just been a long time since I've done C and never ARM
03:19:08linuxstbslimx has done a lot more work on the lcd driver since then though - so we should ask him to submit a new patch.
03:19:25jaebirdI'm not a patch expert either :)
03:19:52jaebirdThat would be great.
03:20:33linuxstbNow's a great time to refresh your memory...
03:21:36linuxstbJust download the Rockbox source code, set yourself up a development environment (see the Documentation section on the Rockbox website), and compile one of the existing iPod builds.
03:21:44linuxstbYou'll then be ready to make a 4g build... :) twisted my arm...already got the code, looking at the docs
03:22:12mikearthurhows the ipod dev going? don't have one, but just interested in how you guys are going
03:22:46linuxstbRockbox is running well (with audio) on the Nano and 4G Color. It runs without audio on the 5G (Video).
03:23:00JdGordon:'( whyt he heck am i getting "undefined reference to '_get_colour'" ?? ive added the correct . h files..
03:23:01linuxstbBut there is still lots of work to do on the Nano and 4G Color ports.
03:23:06mikearthurooo, I didn't know you were getting it running on the nano :D
03:23:13mikearthurhows the ipod port looking compared to the iriver?
03:23:36mikearthurif archos is 100% (ie, being released), where would you say ipod and iriver are?
03:23:47jaebirddoes anyone know if the ubuntu breezy packages suffice?
03:24:29mikearthurjaebird: don't think so, they are gcc 4.0
03:24:49imphasingYou can get a different version of gcc on ubuntu if you specify
03:24:51linuxstbmikearthur: The iriver is probably about 90% there in terms of a first release, the Nano and 4G Color are about 50% there. But a 50% strength Rockbox is still usable...
03:25:16linuxstb(just my opinions)
03:25:16mikearthuryeh, I've been using the iRiver for a while
03:25:22mikearthurI'm convinced its currently the best player on the market, with rockbox support, purely for the gapless
03:25:28mikearthur(I listen to a lot of prog)
03:25:38mikearthurmade a hardware mod to my iriver remote the other day
03:25:44linuxstbjaebird: Don't try and use pre-packaged cross-compilers - they _always_ cause people problems. Just download the source to gcc-4.0.2 and binutils-2.16 and compile them yourself. It's very straightforward.
03:25:50mikearthurrunning over it with my chair :(
03:26:16Paul_The_Nerdmikearthur: GCC4.0 is actually what you use for Rockbox on iPod anyway
03:26:42linuxstbBut as I said 30 minutes ago, it's time for bed. Goodnight.
03:26:52Paul_The_NerdYeah, and actually *sleep* this time.
03:27:05mikearthurlinuxstb: does or should the ipod currently/eventually be able to support gapless?
03:27:14Paul_The_Nerdmikearthur: It currently does.
03:27:28mikearthurI want an ipod now :(
03:28:01Paul_The_NerdThere are a few small glitches to it, I hear from others, but those are glitches, rather than hardware limitations.
03:28:11DreamTactix291i'm still happy with my H140. but the ipod doesn't suck like it used to anymore thanks to rockbox
03:28:51mikearthurI vaguely remember reading something saying flashing an ipod is a lot less dangerous, as you can just restore the old firmware with the apple software
03:28:51mikearthuris that the case?
03:29:15DreamTactix291i believe so
03:29:48Paul_The_Nerdmikearthur: You don't actually *flash* an ipod anyway
03:29:57DreamTactix291isn't it a small partition on the drive?
03:30:00Paul_The_NerdThere's a hidden HD partition, basically, and you just rewrite that
03:30:45Paul_The_NerdThe apple firmware restore utility can rebuild the HD's state to "factory new" so basically, you're safe from lasting harm. And there is a firmware Disk Mode, that never gets touched because it's actually in firmware.
03:31:57DreamTactix291as much as i hate to say it that's a more elegant solution than flashing EEPROM (IIRC)
03:32:23DreamTactix291lot easier to brick an H1xx or H3xx
03:33:16mikearthurpresumably that means some of the disk access and stuff doesn't need to be written for the iPod, as its already in firmware
03:36:15Paul_The_NerdWell, currently when a USB cable is plugged in, it reboots into the hardware "Disk Mode"
03:36:51Paul_The_NerdBut although the easiest solution, I don't think it's the best one in the long term, since at least on Nano "Disk Mode" is noticeably slow.
03:41:36 Quit PaulJ (".")
03:46:30 Part midgey34
03:48:53BHSPitLappydoes it act on a usb1.1 interface or something in hardware diskmode?
03:49:14Paul_The_NerdI'm really not sure.
03:49:40BHSPitLappysomebody lsusb the thing
03:49:48 Join Midgey31 [0] (
03:50:01Paul_The_NerdDidn't bother to check. I applied the patch to load Retail OS when USB is inserted, and have just been using that the whole time
03:58:32 Quit hardeep ("BitchX: now with Olestra!")
04:00:32jaebirdOk. got the toolchain built and a 4g version built!
04:00:55jaebirdNow I just got to get the lcd changes from slimx 3g patch...
04:05:58 Join Lost-ash [0] (
04:05:59 Quit ashridah ("Leaving")
04:09:36 Join Midgey34 [0] (
04:10:41 Nick Lost-ash is now known as ashridah (
04:11:17BHSPitLappyaww, I see my improved h1xx icons never got used on the daily table
04:14:48JdGordonhmm. do i need to do anything to get new files compiled and their functions shared (other that including the nescacery .h) ??
04:20:19BHSPitLappywhat wps's or otherwise thematic things have been made for the nano yet?
04:21:30Paul_The_NerdJust the Boxes one
04:21:54Paul_The_NerdSince I like text based WPSes, I personally just use a couple interesting BG images
04:23:22Paul_The_NerdI've found that filling the whitespace doesn't bother me, like the complicated themes do:
04:25:14JdGordon:'( any devs here?
04:27:48 Quit Midgey31 (Read error: 110 (Connection timed out))
04:28:56BHSPitLappyPaul_The_Nerd: does edison still show up flying through the stars? :P
04:29:01Paul_The_NerdSadly, no.
04:29:10BHSPitLappythat needs to be toggle-able ;)
04:29:14Paul_The_NerdIt's tempting to revert the bit in plugins.c for me.
04:32:23jaebirdSilly me...thinking the patch on current CVS would apply cleanly to slimx Jan 22 patch
04:32:47jaebirdI'm going to have to do this differently...
04:36:37ashridahJdGordon: this a plugin or something else?
04:36:56BHSPitLappyPaul_The_Nerd: are the menu backdrops set in .cfg's?
04:37:06JdGordonno, colour picker for text colour
04:37:25Paul_The_NerdBHSPitLappy: You just browse to a BMP, long hold to pop up the context menu, and choose "Set as Backdrop"
04:37:40BHSPitLappyhow about wps?
04:37:43Paul_The_NerdIf the backdrop is in .rockbox/backdrops (I think that's the name of the folder) it'll be saved through reboots
04:37:48BHSPitLappyyou just set an image, and the text displays in front?
04:38:06Paul_The_NerdIn WPS, you use the %X|image.bmp| tag (capital X) to set that WPS' backdrop. If one isn't set, it uses the global backdrop
04:38:11ashridahJdGordon: hm. well, afaik, it needs to be in an extra text file in the directory it's in
04:39:19JdGordonoh, bugger.... i missed SOURCES.. added it to FILES instead :p
04:40:02BHSPitLappyerm, where are we exactly on text color?
04:40:23Paul_The_NerdWe're "It's not settable yet"
04:40:43JdGordonlinusN wanted a proper colour chooser thingy instead of what the h300 opt. patch has.. and ive been bored today
04:40:52JdGordon... for the hour ive actually been at my computer
04:45:53BHSPitLappyPaul_The_Nerd: only bmp as backdrop, right? no png support?
04:46:16ashridahpng decompression would be seriously taxing.
04:46:23Paul_The_NerdYeah, only BMP, and only at the screen resolution.
04:46:37ashridahit'd be more intelligent to add png support to some kind of rockboxify tool
04:47:01ashridahsince they're thinking of reworking them into one single file containing layout+graphics anyway, so far as i gather
04:47:20Paul_The_NerdOh, was the preprocessor method decided for?
04:47:36ashridahit's been discussed. i don't know if it's been confirmed
04:48:12Paul_The_NerdLast I'd heard the favoured method involved the .wps remaining text, and the bitmaps being in less files (or one) and using the bitmap_part stuff
04:49:21JdGordoncan any1 send me apps/misc.c ?? its taking too long to dl the source from the site
04:49:33BHSPitLappyhey, once i've initiated the WPS, how do I get back to file browser?
04:50:41Paul_The_NerdSelect button
04:51:20BHSPitLappyso smart am I
04:51:49BHSPitLappyerr, how can I browse to the .rockbox folder? (if I even can)
04:52:00JdGordonu need to show all files
04:52:16 Join gantrixx [0] (
04:53:08gantrixxok guys, so I'm placing the order today. Which would you recommend iAudio X5 or the iRiver H340? Which has a better user interface?
04:53:32Paul_The_Nerdgantrixx: By user interface, do you mean "buttons" or "what the screen shows"?
04:53:51BHSPitLappyoh, why can't it be both :)
04:54:11gantrixxthe buttons on the mpio suck and sodoes the graphical interface
04:54:12Paul_The_NerdWell, I was assuming since he was asking in Rockbox he planned to install it
04:54:21Paul_The_NerdWhich means "what the screen shows" becomes relatively immaterial
04:54:32gantrixxassume I will not load rockbox
04:54:48gantrixxlet's just talk apples to apples
04:54:52gantrixx....bad pun
04:55:05BHSPitLappyespecially bad.
04:55:17Paul_The_NerdWell, the X5 is more "modern" so I would *assume* it was more featureful. I believe it supports FLAC, no?
04:55:25CassandraHmm. This is weird. I don't seem to be able to boot the Apple firmware by holding down menu any more.
04:56:10JdGordonashridah: have u got a zip of the latest sources? the rb site is way to slow atm
04:56:11gantrixxmostly I only listen to my media player in my car
04:56:19*JdGordon hasnt got cvs working..
04:56:28gantrixxfor the rest, I stream it to my computer at work from my home mp3 server
04:56:38ashridahJdGordon: uh, i'm on dialup, i really doubt i'm faster :)
04:56:46JdGordonhaha ok
04:57:16Paul_The_NerdCassandra: Did you change something?
04:57:29CassandraUpdated to the latest CVS.
04:57:39CassandraBut that's a function of the bootloader, I believe.
04:57:41Paul_The_Nerdgantrixx: Well, I'd suggest looking at screenshots then and just see which looks best for car visibility
04:57:50Paul_The_NerdCassandra: Yes it is.
04:58:12CassandraWeirdly it boots Apple FW if I try to boot iPodlinux.
04:58:39CassandraStill can't get the darn foobar generated iTunesDB to play.
05:00:41Paul_The_NerdThe Nanos probably use a newer version of iTunesDB
05:01:26 Join Rob2222_ [0] (
05:02:19CassandraI guess so.
05:02:38*Paul_The_Nerd shrugs
05:02:51CassandraI tried using foo_pods 4.7 / 4.8 compabillity setting as well, but that doesn't work either.
05:02:58Paul_The_NerdI couldn't get iPodLinux to see my iTunesControl stuff properly, but I didn't try too hard.
05:03:07CassandraGuess it's a case of wait for the next version.
05:05:50CassandraI don't really care for myself. It's just I want to be able to tell my iPod loving friends that yes, they can still use the Apple firmware and Rockbox.
05:05:59***Saving seen data "./dancer.seen"
05:06:19CassandraIt'll be tricky to get them convinced to try it otherwise
05:07:39Paul_The_NerdI don't have any iPod loving friends, or I'd probably have the same concern.
05:07:48 Quit ehntoo (Remote closed the connection)
05:07:54Paul_The_NerdI have several "Irrational hatred of all things Apple" friends, because they somehow think that makes them better geeks.
05:08:43XavierGrThey do the right thing! ;P
05:09:19XavierGriPod (without Rockbox) is a useless trend. (IMHO)
05:10:07Paul_The_NerdI dunno, the smart playlists seem like a fairly fun feature, though not one I'm likely to use.
05:10:31CassandraI never get that whole my OS/hardware/thing with shiny knobs on is better than yours thing.
05:10:40CassandraI mean most things are good for something.
05:10:45Cassandra(Except IRIX)
05:11:16JdGordonashridah: does the bleedineg source compile? or have i broken something here/>
05:11:34ashridahJdGordon: i have no idea
05:11:41*ashridah isn't overly close to a compiler right now
05:11:58CassandraIt compiled about 6 hours ago.
05:12:25JdGordononplay.c is killing it...
05:12:27CassandraBuilds are all green too.
05:12:43CassandraI think there's something up with your setup.
05:12:44 Quit mikearthur (Read error: 104 (Connection reset by peer))
05:13:02ashridahJdGordon: what's the specific error?
05:13:03 Join mikearthur [0] (
05:13:17JdGordononplay.c: In function `set_backdrop':
05:13:17JdGordononplay.c:459: error: `LANG_BACKDROP_LOADED' undeclared (first use in this functi
05:13:17DBUGEnqueued KICK JdGordon
05:13:17JdGordononplay.c:459: error: (Each undeclared identifier is reported only once
05:13:17JdGordononplay.c:459: error: for each function it appears in.)
05:13:18***Alert Mode level 1
05:13:18JdGordononplay.c:464: error: `LANG_BACKDROP_FAILED' undeclared (first use in this functi
05:13:22JdGordononplay.c: In function `onplay':
05:13:24JdGordononplay.c:611: error: `LANG_SET_AS_BACKDROP' undeclared (first use in this functi
05:13:28JdGordonmake[1]: *** [/home/guest/win32sim/apps/onplay.o] Error 1
05:13:30JdGordonmake: *** [all] Error 2
05:13:33JdGordonsoz.. wasnt supposed ot be tghat many lines
05:13:59ashridahJdGordon: and can you locate where LANG_BACKDROP_LOADED is supposed to be declared?
05:14:47CassandraSounds like your english.lang is out of date.
05:14:48 Quit foo_ (SendQ exceeded)
05:15:28JdGordonah, make clean && make fixed it
05:15:46CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
05:15:46*JdGordon thought i did it already
05:15:47 Quit ts-x ("CGI:IRC (EOF)")
05:15:55CassandraIt's wise to do a make clean after you update cvs.
05:17:11BHSPitLappythat was quite a spectacular flood, JdGordon :P
05:18:47BHSPitLappyanyone know what time the dailies are built
05:19:01 Quit Rob2222 (Read error: 110 (Connection timed out))
05:19:08CassandraIn a couple of hours time.
05:19:33CassandraNext daily will almost certainly be the same as the current bleeding edge.
05:19:55Cassandra6:30 GMT or something. I don't remember the exact time.
05:19:55BHSPitLappyI'm wondering if the 2-02 daily contains the backdrop stuff
05:20:03Paul_The_NerdJust get the bleeding edge.
05:20:39CassandraI believe that it isn't in the 2-02 daily.
05:21:10CassandraYou know, at some point I'm going to stop turning on the iPod just to admire the WPS.
05:23:19***Alert Mode OFF
05:29:59*BHSPitLappy now is aware of the existence of bleeding edge builds.
05:30:39BHSPitLappyhey, I know it's probably minor and all, but the colors are f'd up on the splash screen now on my nano
05:30:48Paul_The_NerdThe bleeding edge should fix that
05:31:03BHSPitLappyapparently not.
05:31:26Paul_The_NerdMy builds from CVS do.
05:32:16BHSPitLappyVer. CVS-060602
05:32:19Paul_The_NerdWhen they removed some files from before the bmp2rb bit was added into the make, it actually started using the real splash bmp, as opposed to the one that was in a source file, and since its colors were messed up, it displayed messed up. But the .bmp file has been replaced with a proper one in CVS now, so that's weird.
05:33:13BHSPitLappyalso, i can't make the bmp my backdrop :P so maybe something's wrong with how I got the bleeding build on there
05:33:30Paul_The_NerdDo you have rockbox.ipod in .rockbox or in the root?
05:33:41Paul_The_NerdCheck to see if there's on in .rockbox
05:34:18Paul_The_Nerd <−− This is what you downloaded?
05:35:40Paul_The_NerdInstalling it now
05:35:49 Quit ashridah ("Leaving")
05:35:58BHSPitLappyand when i downloaded, the table was up to 2006-02-03 00:12
05:36:06BHSPitLappywait a tick
05:36:18BHSPitLappymy version on the ipod says 060202 :S
05:36:42Paul_The_NerdLastes bleeding edit is 060202-0134
05:36:47Paul_The_Nerderrr 060203 rather
05:36:54*Paul_The_Nerd can't type
05:36:59Paul_The_NerdMan, I really can't type
05:37:52BHSPitLappytrying again...
05:38:00BHSPitLappyuh oh, hardware diskmode. this could be a while...
05:38:42Paul_The_NerdUgh, forgot that the CVS version uses Hardware Disk Mode. Ugh.
05:39:17BHSPitLappydid you modify the behavior or something
05:40:35JdGordongood news everybody...
05:40:49Paul_The_NerdThere was a patch for it to default to RetailOS, as there was some debate over which way it should go
05:40:54Paul_The_NerdGood new?
05:41:19JdGordonmy colour chooser thingy works... just got to pretty it up a bit and someone else can make it useable :D
05:43:12Paul_The_NerdHow do you mean "usable"?
05:43:33JdGordonlike actually get the setting the user wants to be the text olour
05:43:40JdGordonor bg col.. or whatever
05:43:47BHSPitLappyah, Paul_The_Nerd...must have been some error in copying the first time
05:44:59BHSPitLappyPaul_The_Nerd: are there bleeding edge sims?
05:45:30Paul_The_NerdAt least, not as far as I know
05:45:44Paul_The_NerdI mean, technically speaking, you can make one by using CVS to get the source and simply building it. :-P
05:45:50BHSPitLappywell, they're built, right
05:46:00BHSPitLappyPaul_The_Nerd: that's what I'm doing now
05:46:02Paul_The_NerdI'm pretty sure the sims aren't built.
05:46:11BHSPitLappyPaul_The_Nerd: look at the build table though...
05:46:18BHSPitLappythey're all built for errors/warnings
05:46:28Paul_The_NerdWell yeah, sorry, aren't archived as binaries
05:47:47Paul_The_NerdSo, you can do it the hard way. :-P
05:48:27Midgey34for win32: download rockboxui.exe and
05:48:47Midgey34unzip to the /archos directory
05:49:54Midgey34you won't have sound though, its disabled in those builds
05:51:21BHSPitLappydoes rockbox have a sleep timer or has my battery been dyinh
05:51:31JdGordonthere is a sleep timer
05:58:03 Quit mikearthur (No route to host)
06:12:29 Join ashridah [0] (
06:14:34 Join DangerousDan [0] (
06:27:46 Quit RotAtoR ("zzzzzz")
06:30:36 Quit DangerousDan (Read error: 104 (Connection reset by peer))
06:32:01 Quit DJ_TCK (Read error: 104 (Connection reset by peer))
06:32:58 Quit imphasing (Read error: 104 (Connection reset by peer))
06:35:33*JdGordon is back....
06:35:54JdGordonwhat does every1 think of
06:37:02ashridahheh. monash. pff. :)
06:37:21JdGordon:D where u?
06:38:07Paul_The_NerdJdGordon: Good enough fro me. :)
06:38:15ashridahwhy is only one of the sliders coloured?
06:38:39JdGordoncoz thats the one that is being fiddled with.
06:43:45BHSPitMonkeyJdGordon: heh, I'm trying to make something just like that in iPL
06:44:00BHSPitMonkeybut not to pick a font color, just display the color
06:44:22JdGordonok.. this will be a widgety thingy so it can be used for other things...
06:44:28JdGordonwhy ipl?
06:45:13JdGordonisnt rb good enough for you....???? </sarcasm>
06:45:38BHSPitMonkeyiPL came first to me :D
06:46:29BHSPitMonkeydo you think you could make your plugin display the hex code somewhere?
06:46:57JdGordonye, its gonna... as soon as i fix 1 little freeking line
06:47:28BHSPitMonkeyit'd also be cool to show each of the color values at the end of their slider
06:47:38BHSPitMonkeybut not as necessary if the hex is there anyway
06:48:37JdGordon.. thats what i was gonna show.. not the combined val
06:48:46JdGordoncoz of the pain with rgb or bgr lcd..
06:55:58JdGordonwhats the generic cancel button on the ipoo?
06:59:02JdGordononly ipod and h300 have colour screens right?
06:59:11JdGordoni mean, no other targets?
06:59:39Paul_The_NerdAt the moment, yes
07:00:19BHSPitMonkey176x132, 220x176, and 320x240 are the only color screen sizes
07:03:38JdGordonok.. wierd bug... i can hold right and then press on and it works fine.. but iff i press on and then hold that and right it crashes??? wtf (the code is case SLIDER_OK|BUTTON_RIGHT:
07:03:39JdGordon rgb_val[slider] = (rgb_val[slider]+0xf)%0xff;
07:03:39JdGordon break;)
07:06:00***Saving seen data "./dancer.seen"
07:09:23 Join deeku [0] (
07:09:53deekuhey, im trying to set up a rockbox bootloader for an iriver h300
07:10:46BHSPitMonkeyhey, I'm stephen
07:10:53BHSPitMonkeynice to meet you, trying
07:11:56deekui can't seem to compile the tools descramble/mkboot/scramble from a source tarball
07:12:23deekuon a linux box
07:12:27BHSPitMonkeydo you have to compile anything?
07:14:01Rob2222_good morning
07:14:05 Nick Rob2222_ is now known as Rob2222 (
07:15:24Paul_The_Nerddeeku: What kind of errors are you getting?
07:15:54deekuit seems to be telling me to use an m68k compiler or something... im not sure
07:16:14deekuthe stuff it looks for in my path, that arent there
07:16:38Paul_The_NerdOdd. I don't thing scramble/descramble/mkboot should use the cross compiler...
07:16:54deekuit might be a problem with the bleeding edge source... im trying an older one now
07:16:56ashridahPaul_The_Nerd: they don't, but the configure script looks for the cross compiler anyway, iirc.
07:17:24Paul_The_Nerddeeku: Do you have the cross compiler?
07:17:32deekuthe cross compiler?
07:17:43Paul_The_NerdYou need the m68k compiler to build rockbox itself, and the bootloader image
07:17:43deekui have gcc.. im not really sure
07:18:04deekuoh i see. but what if i just need the tools scramble/descramble/mkboot
07:18:04ashridahhe definently shouldn't be compiling the bootloader himself.
07:18:22ashridahinserting it into the .hex sure, using a built one, no.
07:18:22deekuno im not trying to compile the bootloader
07:18:40ashridahdeeku: what command are you running when it fails?
07:18:44deekuyeah i need the tools to set up my iriver .hex bootloader
07:18:47deekuhold on im trying again
07:18:59JdGordonhmm.. ok.. ON doesnt to repeat on the player.
07:19:17ashridahdeeku: got wine handy? fwpatcher.exe works in that if you can't be bothered wrestling with it.
07:19:24 Join B4gder [0] (
07:19:49deekuok ill try wine
07:20:00deekuis there an option for an apple box?
07:20:04ashridahdeeku: i *think* you can build the tools if you build the simulator, i'm not sure if it includes all of them tho
07:20:29deekuis the fwpatcher.exe tool graphical?
07:20:33deekuor command line only
07:20:48B4gderthere are command line tools too
07:20:52ashridahie, make a directory, go into it, run ../tools/configure, select 10, then simulator, then sdl, and run make in that. even if the sim itself doesn't build for some reason, the tools might
07:20:53B4gderjust not as fancy
07:20:54deekuwhere do i get those?
07:21:06B4gderyou build them from source
07:21:08deekuok ill try that
07:21:33ashridahdeeku: which is what he's attempting to do, but he isn't too familiar with rockbox's build system, so what he's doign currently is asking for a cross compiler
07:21:46ashridahB4gder: even
07:22:28deekuyea i get what's goin on now, thanks for the help :)
07:22:50B4gderyou don't actually need a cross-compiler just for that
07:22:59B4gdersince the bootloader is available for download
07:23:04deekuok yea it just built successfuly
07:23:07B4gderand the patcher tools are for host
07:25:57Rob2222 <= what do i need for the sim? the .exe always says "no .rockbox dir" but i have it there
07:26:01JdGordonright.. i tihnk im done...
07:26:15BHSPitLappyRob2222: unzip the zip?
07:26:20B4gderRob2222: you did 'make install' ?
07:26:39Rob2222i just downloaded the .zip and the .exe
07:26:51BHSPitLappyRob2222: the .rockbox folder goes inside a folder called archos
07:27:10B4gder'archos' being the simulated root
07:27:16BHSPitLappywhich represents the root of the...damnit
07:27:16Rob2222thats it
07:28:42Rob2222hey, that sim is cool
07:28:55BHSPitLappyI hate the keymap.
07:28:59BHSPitLappy(for the ipod one)
07:29:12Rob2222can it play mp3s ? :D
07:29:26B4gderit can
07:29:37B4gderbut for some reason that is not enabled by default
07:31:24JdGordonum... whats button_up and _down on the ipod? (or equivilants?)
07:31:52Paul_The_NerdEither Menu and Play, or the scroll left and right
07:34:25Rob2222i deleted _temp_codec.dll now i get a error message every start
07:34:39JdGordonhmm.. good old apple not putting enough buttons on... i need up/down, left/right, ok and cancel... any suggestions?
07:35:07Paul_The_NerdSelect for OK
07:35:15Paul_The_NerdMenu for cancel
07:35:24Paul_The_NerdScroll the wheel left and right to move the sliders
07:35:29BHSPitLappyRob2222: undelete it?
07:35:34Paul_The_NerdAnd |<< and >>| for selecting between them
07:35:58Rob2222ah, dumb me. dont use mk68k binaries (codecs) at win32
07:36:47B4gdernope, no emulator only simulator
07:37:20Rob2222is the sim build at cvs changes, too?
07:37:41JdGordonPaul_The_Nerd: can select be used in combination with other keys?
07:37:49Paul_The_NerdYes, it can
07:37:57B4gderRob2222: yes, all the ones mentioned in the cvs build table are
07:37:58Paul_The_NerdSome of the plugins use Select+Menu to quit
07:37:59JdGordonall good then
07:39:12Paul_The_NerdLookin' forward to a .diff or .patch soon. ;-)
07:39:30JdGordonor a good ol' .c :D
07:39:49Rob2222dumb question: is the rwps file for the remote?
07:39:55Paul_The_NerdRob2222: Yes
07:40:17Rob2222omg. i use rb since... 2 month? and didnt came to this idea
07:40:32Rob2222i thought r was for archos recorder
07:40:40JdGordonhehe.. silly bugger
07:44:01 Quit Thus0 (Read error: 104 (Connection reset by peer))
07:49:21 Join LinusN [0] (
07:49:25Rob2222damn that sim crashes and crashs
07:50:07Rob2222good morning LinusN
07:52:02JdGordonhey LinusN
07:53:55 Join Siku [0] (
07:56:32JdGordonkeys in the ipod sim are retarted D:
07:56:56Paul_The_NerdThey bein' difficult?
07:57:16JdGordonit just feels wierd... 8 is down and 2 is up..
07:57:37JdGordonanywho.. im done
07:58:03JdGordonbut its not really a workable patch... someone needs to figure out how to deal with the colour it returns
07:58:23JdGordonbut it works fine...
07:58:35Rob2222that shouldnt be that hard, or?
07:58:50JdGordonwell.. i dont know how ppl want to deal with it.. so ill leave it
07:58:54LinusNJdGordon: ah, nice
07:59:17Paul_The_NerdSo right now it doesn't actually change the color, it just lets you *pick* one, and then returns it, and nothing is done with the return?
07:59:41LinusNJdGordon: how does it handle the different fontsizes?
07:59:57JdGordonmight not like the very large ones tho..
08:00:06LinusNhehe probably not
08:00:26LinusNdo the sliders shrink with larger fonts?
08:00:29Paul_The_NerdDoes it fit on the Nano's screen properly?
08:01:02Rob2222that would be nice if it can be used.
08:01:44JdGordonwell... actually.. it sort of can be used now.. untill something else changes the foreground colour... but thats not really ideal
08:02:14JdGordonhmm.. sf is down? shall i email it or forum it?
08:02:22LinusNJdGordon: where is the patch?
08:02:55JdGordonlike i said.. its only the frontend.. u need to put it how u want it... its 1 .c atm
08:03:23JdGordonwell.. and the .h if u really want 1 funciton def..
08:03:38Paul_The_NerdSo, something needs to be in place to preserve the color, so that when a plugin changes it, it'll restore properly?
08:05:14Paul_The_NerdWell, the only thing that changes FG color are it, and plugins, right?
08:05:15Rob2222JdGordon: Could you make some selectable presets there, too?
08:05:24Rob2222maybe black, white, green ...
08:05:38Paul_The_NerdRob2222: All presets do though, is clutter...
08:06:49Rob2222maybe only 5, or 6 ... little squares between the sliders and the color field
08:06:59Rob2222but well, was just an idea
08:08:12Rob2222black white red yellow green blue ;)
08:08:52Paul_The_NerdJdGordon: Care to share your .c?
08:09:01Paul_The_NerdMaybe put it on the webspace for the moment?
08:09:10JdGordonit could be done.. but it will clutter it up.. and not look good if larger fonts are used... or the screen is small
08:09:12JdGordonsoz.. 1 min
08:10:09JdGordonno laughing at the horrible code
08:12:39 Join moda [0] (
08:12:53Paul_The_NerdGonna try it out then, and maybe see if I can add some handling around the plugin loader. :)
08:13:02Paul_The_NerdThat shouldn't exactly be "hard" ;-)
08:13:03modalol, tried to play goldeneye on my iriver and completely crashed it :D
08:13:53LinusNPaul_The_Nerd: isn't that handler by the patch in the optimized version?
08:14:26Paul_The_NerdLinusN: Dunno, I don't have an h300, so I've never even looked at it. If it is, then I guess I don't need to do anything. I bet they have a WPS tag for FG color too, don't they?
08:14:46LinusNi think so
08:15:14LinusNnot sure though, i've never tried the optimized version
08:15:15JdGordoncomment out line 156 and the colour will stay when u exit that screen...
08:15:22^BeN^LinusN, can you do a transparent drawmode?
08:15:34^BeN^so we can use it in the plugins too?
08:15:37 Nick ^BeN^ is now known as Paprica (
08:15:49LinusNPaprica: huh?
08:16:00Papricafor text..
08:16:06Papricanot for an image...
08:16:46LinusNi thought text could be transparent already
08:16:48Paul_The_NerdLinusN: Okay, after trying out JdGordon's thing on Nano, I'll look at that patch 'n see what it's got, and maybe merge the two.
08:17:31Papricawhich drawmode
08:18:17LinusN*should* work
08:18:27Papricatry it
08:18:34Paul_The_NerdJdGordon: Do I need to do anything other than just include it in the Source file?
08:19:01JdGordonPaul_The_Nerd: if u do merge the two.. can u change ln 63 to use the current fg colour instead of the default?
08:19:17JdGordonjust declare the get_colour functino somehwere
08:19:36JdGordonor use
08:20:00LinusNi think the slider rendering should move to apps/recorder/widgets.h
08:20:08LinusN.c even
08:20:18 Quit moda ()
08:20:28LinusNand maube make the slider look more like the scroll- and progress bar
08:20:57JdGordonif u want, i cant play with it any more today
08:22:03LinusNi think i'd like a better visual representation of the currently selected slider, live inverting the text
08:22:19PapricaLinusN, it works! thanks..
08:22:22LinusN(what's with my typing today?)
08:22:27LinusNPaprica: goodie
08:22:39JdGordoninvert the text how?
08:23:13LinusNlike the browser cursor is when you select "inverse bar"
08:24:36LinusNfrom put_cursorxy in menu.c:
08:24:42LinusN lcd_set_drawmode(DRMODE_SOLID|DRMODE_INVERSEVID);
08:24:42LinusN lcd_fillrect (xpos, ypos, 4, 8);
08:24:42LinusN lcd_set_drawmode(DRMODE_SOLID);
08:25:28LinusNgotta run, cu later
08:25:31 Part LinusN
08:28:07JdGordonhmm... how does the scrollbar work? maybe that widget could be used instead of the current slider?
08:30:18 Quit deeku ("leaving")
08:31:23Paul_The_NerdJdGordon: What are you supposed to pass get_colour?
08:31:59Paul_The_NerdI mean, why does it let you pass it pointers for the RGB values, if it changes the foreground itself?
08:32:58 Quit ashridah (
08:32:58 Quit Rob2222 (
08:32:58 Quit Midgey34 (
08:32:58 Quit lostlogic (
08:32:58 Quit [San] (
08:32:58 Quit JdGordon (
08:32:58 Quit Paprica (
08:32:58 Quit jjoao (
08:32:58 Quit DreamTactix291 (
08:32:58 Quit Rick (
08:32:58 Quit Rob- (
08:32:58 Quit ShyK (
08:32:58 Quit Maxime (
08:32:58 Quit beeble (
08:32:58 Quit godzirra_ (
08:32:58 Quit preglow (
08:32:58 Quit dpassen1 (
08:32:58 Quit crwl (
08:32:58 Quit radsaq (
08:32:58 Quit pabs (
08:33:22NJoinashridah [0] (
08:33:22NJoinRob2222 [0] (
08:33:22NJoinMidgey34 [0] (
08:33:22NJoinlostlogic [0] (
08:33:22NJoin[San] [0] (
08:33:22NJoinJdGordon [0] (
08:33:22NJoinPaprica [0] (
08:33:22NJoinjjoao [0] (n=joao@
08:33:22NJoinDreamTactix291 [0] (
08:33:22 Join Rick [0] (i=rick@unaffiliated/Rick)
08:33:22NJoinRob- [0] (
08:33:22NJoinShyK [0] (i=Shy@
08:33:22NJoinMaxime [0] (
08:33:22NJoinbeeble [0] (
08:33:22NJoingodzirra_ [0] (
08:33:22NJoinpreglow [0] (
08:33:22NJoindpassen1 [0] (
08:33:22NJoincrwl [0] (
08:33:22NJoinradsaq [0] (n=radsaq@
08:33:22NJoinpabs [0] (
08:33:25***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
08:33:54JdGordoni knew they couldnt resist us :D
08:34:32Paul_The_NerdJdGordon: Did you have an intent for why it has the red green and blue pointers?
08:34:54PapricaPaul_The_Nerd, scrollbar(int x, int y, int width, int height, int items, int min_shown,
08:34:54Papricaint max_shown, int orientation)
08:34:58JdGordonye, so if some1 decided they wanted them.. also it gives the default colour
08:35:00Papricawhat is the problem?
08:35:20JdGordonitems is the current position between min and max?
08:36:40Papricabalh dont know what is the items
08:39:04JdGordongah, it looks like its a log scale or somethng...
08:50:04Paul_The_NerdUgh, can't work on it now anyway. Gotta go.
08:50:05 Part Paul_The_Nerd
08:53:42 Join ender` [0] (i=ychat@
08:54:48 Join Musicmad [0] (
08:54:57 Join banan_ [0] (
08:55:56 Join damaki [0] (
08:56:32 Quit foo__ (SendQ exceeded)
08:57:00Papricasomeone can help me with read_bmp_file?
08:57:57Slasherimorning :) just waiting for train to helsinki :)
08:58:14Papricahi =]
09:01:12 Join imphasing [0] (
09:03:24 Join banan__ [0] (
09:03:49BHSPitMonkeywhat is AR+RANLIB during a build?
09:03:57BHSPitMonkeyit seems to take forever / get stuck sometimes
09:04:20B4gderbelieve it or not, but it runs ar and ranlib! ;-)
09:04:23imphasingranlib would set up the libraries for your system, so it can use them..
09:05:11*BHSPitMonkey goes into a rivalrous glaring contest with imphasing
09:06:03***Saving seen data "./dancer.seen"
09:06:41 Quit merbanan (No route to host)
09:06:44*imphasing doesn't blink
09:06:53*imphasing ← has the skills
09:07:00*BHSPitMonkey also has "the skills"
09:07:18*BHSPitMonkey throws his nano directly at imphasing's eye, breaking a blood vessel
09:07:44imphasingchristi/misc/iPod-Rockbox.JPG"> #!!!
09:08:11BHSPitMonkeyimphasing: yup, that's why my wps looks like atm
09:08:14B4gdervery dark picture ;-)
09:08:56 Quit damaki_ (Read error: 110 (Connection timed out))
09:08:56imphasingI feel like putting rockbox on my ipod..
09:09:03B4gdergo go go
09:09:18imphasingI wouldn't mind piddling around with the source though
09:09:31B4gderslimx seems to be progressing on the 3g front
09:09:59imphasingI've been thinking of buying a new ipod anyway though..
09:10:14BHSPitMonkeystill had some horizontal lcd packing business to sort out the last time I checked in
09:10:28BHSPitMonkeyfor the greyscale's
09:11:36imphasingThe CVS was down last time I tried getting a source tree..
09:11:46imphasingI wouldn't mind a just ipod source tree though.
09:11:59imphasingYeah, no response from server or some such
09:12:22B4gderme knowingly, the CVS has not been down for the last 170 days
09:12:37imphasingMust have been me.
09:12:47B4gderyeah, or some silly network problem somewhere or so
09:12:50 Join merbanan [0] (
09:13:32imphasingwow, fast CVS..
09:14:08BHSPitLappyespecially from overseas...
09:16:44B4gderwell, we had the CVS on sourceforge before that and everything is better than SF :-O
09:16:54imphasingHeh, true
09:17:01imphasingthey have very slow CVSs
09:17:05imphasingand often unreliable..
09:17:32imphasingIs there a build doc for building the ipod port?
09:17:48B4gderimphasing: the actual build process is the same for all targets
09:18:13imphasingIs there a doc for that?
09:18:15B4gdercreate build dir, run configure, run make, run make zip
09:18:19imphasingI have an ARM toolchain already
09:19:36 Quit banan_ (Connection timed out)
09:19:37BHSPitLappyimphasing: to configure the bootloader, you need 4.x
09:19:39preglowlinuxstb: didn't you say you commited some alac stuff?
09:19:50*preglow cries for the missing clz opcode zsh keeps resetting my PATH
09:20:30BHSPitLappyimphasing: you know that every time there's a commit, builds are made/hosted, right
09:20:55 Join colin_ [0] (
09:20:58imphasingI want to compile though!
09:21:01imphasingThat's half the fun!
09:21:08B4gderthat's the spirit!
09:21:15BHSPitLappyI compile fine...from cygwin
09:21:21BHSPitLappywhen I try on my linux box it goes crazy
09:21:25imphasingIf you can't edit some source and ruin everything, then you're doing it all wrong.
09:22:00ashridahdamnit. wish i could remember what email address i used for sourceforge
09:22:16imphasingWow..that was the easiest build process I've ever done..kudos!
09:22:33colin_Hey guys, quick question. I noticed that the Rockbox article over at the hydrogenaudio wiki mentions that 'preliminary support' is available for the iPod mini, but I can't find confirmation on the site. Is this simply bleeding-edge support, or should I remove the mention?
09:22:34*B4gder takes some credit for that
09:22:40JdGordonwhere is the code for drawing the progress bar on the wps?? i cant find it
09:22:46imphasingB4gder: You're a developer?
09:22:48BHSPitLappycolin_: nothing yet
09:22:59B4gderimphasing: yessir
09:23:14BHSPitLappycolin_: slimx is working hard on the greyscale ports, they're getting close
09:23:20colin_BHSPitLappy: Thank you, I'll correct the article appropriately :)
09:23:26colin_I'm eagerly awaiting mini support ;)
09:23:29JdGordonB4gder: do u know how the scrollbar widget works?
09:23:41BHSPitLappycolin_: don't quote me :P I'm not a dev
09:23:51imphasingBHSPitLappy: I might have to peek in on what's going on..
09:23:53 Join damaki_ [0] (
09:24:36imphasingHas anyone had a catastrophe when trying to run the ipod color drivers on a 3G?
09:25:08BHSPitLappyI may not know what I'm talking about, but that sounds ridiculous.
09:25:29imphasingIt probably is.
09:25:50imphasingBut writing to addresses outside of an actual devices range could have unexpected results..
09:25:52BHSPitLappy^ get grepping, lol
09:26:49linuxstbpreglow: Yes, I did commit an alac change. No idea what happened there....
09:27:24B4gderJdGordon: I suggest you check the code that already uses it and see how that works
09:27:25linuxstbpreglow: I've just checked my bash history - seemed I typed "cvs diff" instead of "cvs commit" :) It was very late...
09:27:39JdGordonB4gder: i did.. its wierd.. the ipod's LCD driver really only 285 lines?
09:28:01linuxstbpreglow: Committed now.
09:28:08JdGordonno big deal
09:28:08preglowlinuxstb: if our arm had had the 'clz' instruction, the optimising of the count_leading_zeroes() function would have been easy indeed...
09:28:14linuxstbI know :)
09:28:29ashridahimphasing: it probably shares code with many of the other drives
09:28:41ashridahs/drives/devices/ even
09:28:43B4gderimphasing: the graphics later in Rockbox draws everything in a framebuffer, so the "driver" part is only about transferring that buffer to hw
09:28:47linuxstbimphasing: The high-level lcd code is in driver/lcd-16bit.c
09:29:01imphasingAh, ok.
09:29:31 Quit banan__ (Connection timed out)
09:29:44 Quit Musicmad (Read error: 110 (Connection timed out))
09:29:54imphasingAre the developments for the 3G in the CVS?
09:30:03BHSPitLappydun' think so
09:30:13BHSPitLappyslimx is the gatekeeper and the keymaster
09:30:14colin_linuxstb: I noticed that the Rockbox article on the hydrogenaudio wiki mentions 'preliminary support' for the iPod photo, nano, and mini, but have been unable to find anything confirming mini support on the Rockbox site. Is it just bleeding edge support, or should I remove that comment?
09:30:25linuxstbThere is a patch on Sourceforge, but that's about 12 days old - the person working on it has done a lot more since then.
09:30:26B4gderimphasing: there's a patch in the tracker though
09:30:33imphasingA patch!
09:30:36B4gderach, too slow
09:30:50 Join damaki__ [0] (
09:30:51BHSPitLappyimphasing: yes, this project is just riddled with patches
09:30:54imphasingIs there a way I can get a hold of him?
09:30:57BHSPitLappylike a quilt, rockbox is!
09:31:00linuxstbcolin_: That's a mistake on the HA wiki - Rockbox doesn't support the mini at all at the moment.
09:31:06B4gderimphasing: he pops in here regularly
09:31:08BHSPitLappyimphasing: join #rockbox during business hours :
09:31:12preglowlinuxstb: Zero Count
09:31:15preglowlinuxstb: looks faster to me
09:31:15imphasingI'll just hang around until I see him poop up.
09:31:26colin_linuxstb: Thanks, I'll make the correction :)
09:31:55Paprica[09:58:00] <Paprica> someone can help me with read_bmp_file?
09:32:11preglowat least it might be, it has no branches
09:33:21imphasingArg..sourceforge slowness..
09:34:42linuxstbimphasing: The port to the 3g doesn't even start Rockbox yet - it just makes a start on getting the bootloader to work.
09:35:22linuxstbCurrent status is that the bootloader runs, it can start the retailos, but the ata driver isn't working.
09:35:24BHSPitLappyhaha B4gder
09:35:37BHSPitLappy"Poor Bagder, nobody likes him/her. He/She was attacked 16 times."
09:35:47BHSPitLappyIRC stats sites are fun!
09:36:32imphasinglinuxstb: That's good enough for me.
09:36:53 Quit damaki (Read error: 110 (Connection timed out))
09:36:59imphasingI thought the ata driver for all the ipods was identical though
09:40:22 Join damaki [0] (
09:42:03 Quit damaki_ (Read error: 110 (Connection timed out))
09:43:58linuxstbpreglow: I've put that CLZ function into alac, and I think it's slightly faster, but not that significantly.
09:44:13preglowlinuxstb: i'll optimise it in asm anyway
09:44:15preglowsooner or later
09:44:30B4gderreminder for some fun reading:
09:44:46linuxstbpreglow: Have you searched for an ARM implementation?
09:45:00preglowlinuxstb: no
09:45:04preglowlinuxstb: i just remembered that one
09:45:10preglowB4gder: ahahah, let the lawsuits commence
09:45:48imphasingB4gder: Any software company named "softjock" needs to rethink their business plan..
09:46:10 Join [IDC]Dragon [0] (
09:46:10linuxstbpreglow: Another implementation:
09:46:51preglowvery arm friendly as well
09:47:16[IDC]Dragonlinuxstb: I've seen you mentioning count_leading_zeros(), you know the ARM has an instruction for that?
09:47:20pregloweach of the ors will take one instruction
09:47:25preglow[IDC]Dragon: not ours
09:47:27linuxstb[IDC]Dragon: Not the arm7tdmi
09:47:32preglow[IDC]Dragon: we only have armv4t
09:47:37preglowclz is arm5
09:47:47[IDC]Dragoni forgot
09:47:48preglowapple are cheapskates :/
09:48:07[IDC]Dragonso you dont have MAC, saturating math and all the goodies
09:48:12preglowmac, yes
09:48:15preglowbut not saturating math
09:48:28preglowwe hav mla, smull and smlalk
09:48:31[IDC]Dragonoh well, mac helps
09:48:36preglowit helps tons
09:49:25imphasingNow why would you want to count leading zeros?
09:49:34imphasingAnd what would you be counting them in?
09:49:44[IDC]Dragonok, I'll keep quiet instead of trying unqulified suggestions
09:50:00 Quit JdGordon (Read error: 104 (Connection reset by peer))
09:50:06preglowlinuxstb: btw, that function would be very small in arm asm, each or line is one instruction
09:50:44 Join Lynx_ [0] (
09:54:30 Quit damaki__ (Read error: 110 (Connection timed out))
09:55:50 Quit jjoao (Read error: 110 (Connection timed out))
09:56:23[IDC]Dragonquietness is boring, excuse my curiosity: what's missing for audio on the video ipod?
09:56:35[IDC]Dragonnobody knows the hardware?
09:56:47preglowand no one seems to have tried very hard either
09:57:35[IDC]Dragonso many interesing projects out there I have no time for... :-(
09:58:14BHSPitLappypreglow: don't fret, the boys at iPL will have it in no time!
09:58:16BHSPitLappyheh heh.
09:58:24*BHSPitLappy runs
09:58:30[IDC]Dragonare the 2 CPUs "symmetric", or specialized to one or the other thing?
09:58:36preglow[IDC]Dragon: symmetric
09:58:50preglowtwo more or less exactly equal arms
09:58:53preglowseparate caches
09:59:00BHSPitLappyhow analogous
09:59:07[IDC]Dragonand both can access all the hardware?
09:59:14preglow[IDC]Dragon: afaik
09:59:22preglowthey've got the same memory space
10:00:14 Quit midkay (Read error: 104 (Connection reset by peer))
10:00:32imphasingIt's sad that you can't treat one as a coprocessor, then you could just use ARM's coprocessor instructions..
10:01:00[IDC]Dragonwhich is very cumbersome
10:01:36imphasingBut better than waking up one processor, loading some code into another part of ram, and coaxing the other processor to access that code and start running it..
10:01:37[IDC]Dragonwe never built coprocessors, always used AHB-mapped I/O
10:02:07imphasingAre there MMIO coprocessors?
10:02:16 Join midkay [0] (
10:02:34preglowimphasing: that part really isn't very hard
10:02:55preglowthe hard part is you always keeping in mind that the other cop is running somewhere in the same memory space, possibly touching the stuff you do
10:03:00preglowand the non-coherent caches
10:03:44preglowwe'll no doubt have a couple of _very_ interesting bugs
10:04:04linuxstbpreglow: I'm giving up now. It's hard to tell which of the C routines is fastest - I think it's very close between the one you found, and the one I found. I'll try again with lostlogic's null output plugin (if it's working on ipod)
10:04:08imphasingI wonder if it would be possible to write some functions that would treat the other core as a MMIO coprocessor, then you could just send it an instruction, and it would do it, with no real complications..
10:04:14imphasingI'm probably being irrational though.
10:04:58imphasingMemory Mapped IO
10:05:08linuxstbpreglow: But I still haven't managed flawless ALAC playback - disk reading is still too much for it.
10:05:10preglowimphasing: there's the mailbox
10:05:17preglowimphasing: but that wouldn't be very efficient
10:05:38preglowimphasing: yeah, one processor puts data in a place in memory, and the other one receives an interrupt telling it something arrived
10:05:50imphasingI bet that's slow..
10:05:57preglowlinuxstb: really weird how it can be _that_ slow
10:06:13preglowlinuxstb: how's it work when buffering has been done?
10:06:28linuxstbIt's fine when there is no disk reading - the pcm buffer is kept full.
10:06:42preglowhow long are the frames in alac?
10:06:58linuxstbYou mean in samples?
10:07:05preglowyeah <-
10:07:40 Join LinusN [0] (
10:07:49linuxstbI _think_ it's fixed at 4096.
10:08:11preglowi really think there's got to be some other cause for it
10:08:17preglowit can't possibly be slower than aac
10:08:31preglowtry using the null output thingie
10:08:59linuxstbI will do, but not now.
10:09:01preglowi'd be well-nigh shocked if it turned out to be decoding at slower than realtime
10:09:19linuxstbNo, it's definitely faster than realtime with no disk access.
10:09:26preglowi should commit eq ui before i leave
10:09:29imphasingpreglow: #?
10:09:37BHSPitLappypreglow: screenie?
10:09:52linuxstbPaprica: Still having problems with read_bmp_file ?
10:09:55preglowimphasing: what about it?
10:10:04imphasingThat's probably very inefficient
10:10:12imphasingthat's what you guys were talking about?
10:10:13preglowyes, i think so
10:10:15preglowwhere'd you get it?
10:10:17imphasingfor counting the leading zeros?
10:10:17preglownot very, but a bit
10:10:27imphasingIt's from some newsgroup
10:10:30Papricalinuxstb, yep
10:10:40linuxstbPaprica: What is the problem?
10:10:44preglowit uses no table lookups
10:10:49preglowand only scratch regs
10:10:50preglowit might be fine
10:10:52preglowlinuxstb: try!
10:11:07Papricait doest load the image..
10:11:20imphasingpreglow: heh..
10:11:23Papricai try to use what it load
10:11:27linuxstbpreglow: Can you turn that into a C function for me? My asm skills are still developing.
10:11:28Papricabut it show nothing
10:11:39preglowturning it into c would defeat the purpose
10:11:42preglowit's an arm clz
10:11:48preglowuses conditional instructions extensively
10:11:51imphasingI have another one, if you want to see it.
10:11:53linuxstbI mean inline asm in a C function.
10:12:13preglowgimme a sec, i've got to go arrange stuff for now
10:12:21linuxstbNo rush - I'll be around all day.
10:12:32preglowi won't
10:12:39preglowgoing away for the weekend in an hour
10:12:55linuxstbPaprica: Can you pastebin the relevant parts of your program?
10:13:22 Join Strath [0] (
10:13:23 Quit StrathAFK (Read error: 104 (Connection reset by peer))
10:14:13 Join webguest50 [0] (
10:15:40webguest50I am trying to find the 'Optical' on, off setting in the config file, but I cant find it, I have created a config with it on, (H140)
10:16:10 Part LinusN
10:16:11 Join LinusN [0] (
10:16:30 Part LinusN
10:16:55imphasing #here's one that uses a loop to count them. I'm not really sure about the efficiency of it's probably pretty bad.
10:17:02 Join LinusN [0] (
10:17:57webguest50the config file does switch optical out on, so where does it hide it in the config file, (line number ?) or am I just blind
10:18:08 Join petur [0] (
10:19:36linuxstbwebguest50: Try searchiing for spdif in your config file
10:20:00LinusN"spdif enable"
10:20:15 Nick colin_ is now known as colin_sleep (
10:20:21peturis SF down or is it just my end of the cable?
10:20:29B4gderseems to be down for me too
10:20:32linuxstbIs ice cold?
10:21:07imphasingMe too..
10:21:19Papricaok for me..
10:21:22webguest50well who would have guessed, spdif = optical, :)
10:21:48LinusNwebguest50: nor that far fetched imho
10:22:17webguest50acronum I'm not familiar with obviously
10:22:56midkayLinusN: making the cross compiler now.. so, shall i make two dirs in "build", gcc3 and gcc4, and install them both seperately, one with target=sh-elf and one with target=arm-elf, and it'll work out fine like that?
10:24:01LinusNyes, i use to call the dirs after the target name
10:24:08LinusNmidkay: are you using cygwin?
10:24:13midkayno, linux..
10:24:51midkayhooray :)
10:25:08LinusNbetter have different prefix for them
10:25:30LinusNexample −−target=sh-elf −−prefix=/usr/local/sh
10:25:36imphasingIs the long on the ipod 32 bits?
10:25:54midkayand -sh+arm for gcc4?
10:25:58linuxstbPaprica: I think you should pass FORMAT_NATIVE, not FORMAT_ANY. Does gcc give you any warnings about that code?
10:26:52linuxstbI think you should be passing a pointer to your icons struct to that function - not the struct itself. i.e. "struct bitmap* icons"
10:27:18 Join StrathAFK [0] (
10:27:19 Quit Strath (Read error: 104 (Connection reset by peer))
10:27:53imphasinglinuxstb: #if you still care
10:27:57linuxstbAlso, the memory you are assigning to is a local variable inside that function - so it will be on the stack. You should define img_buf to be a global variable.
10:28:41linuxstbimphasing: Thanks - I've already seen a version the same as that. Your first ARM asm one seems the best so far - I'm waiting for preglow.
10:28:55imphasingDid you try running it?
10:29:02preglowthen you'll have to wait until monday
10:29:08imphasingThere's also the one that used a loop, which probably was alot slower
10:29:38 Quit BHSPitLappy ("Dropped my laptop.")
10:29:44Papricalinuxstb, ok
10:31:29linuxstbpreglow: No problem. I'll start learning.
10:32:31imphasinglinuxstb: If I'm not mistaken, you can just link that single function in with whatever C file you're compiling, so you can call it from there..
10:32:39imphasingor you could use that asm() function I guess..
10:32:44imphasingBot sure how slow that is though
10:34:31 Part webguest50
10:34:47 Quit kkurbjun ("Leaving")
10:35:07 Join Paul_The_Nerd [0] (
10:36:30preglowokok, i've broken the equaliser somehow
10:36:33preglowi'll have to come back to this
10:36:52preglowi probably messed something up when changing the fixed point format of q
10:37:36linuxstbimphasing: Do you have a link to the original source of that first function you posted?
10:38:18imphasingYou mean where I got it?
10:39:27Slasheriah, finally in train.. it was over an hour late :/
10:39:55linuxstbimphasing: Yes, thanks.
10:39:57imphasingThat has a few examples, but from what I saw, the one I pastebin'd looked best.
10:40:35imphasinguses 17 instructions, as opposed to around 20-30 for the others
10:40:47 Join safetydan [0] (n=dan@
10:41:02imphasing MOV R1,R0
10:41:02imphasing MOV r0,#0
10:41:02imphasing.loop MOV r1,r1,lsl#1
10:41:02DBUGEnqueued KICK imphasing
10:41:02imphasing MOVCS pc,r14
10:41:02imphasing ADDCC r0,r0,#1
10:41:02***Alert Mode level 1
10:41:02imphasing BCC loop
10:41:06imphasingLike that one.
10:41:26preglowthat's a naive one
10:41:38imphasingTakes about 35 instructions
10:46:40preglowbut yeah
10:46:46preglowi'm outa here, see you later
10:51:03***Alert Mode OFF
10:52:10 Quit mafj2 (Remote closed the connection)
11:03:46midkaygetting a bmp2rb error on build, any ideas?
11:04:11midkayBMP2RB rockboxlogo.112x37x1.bmp
11:04:11midkaymake[3]: *** [/home/midkay/.local/share/Trash/files/rockbox/build/apps/bitmaps/native/rockboxlogo.112x37x1.c] Error 1
11:04:11midkaymake[2]: *** [all] Error 2
11:04:18imphasingGah, building the bootloader for ipod looks like it sends make into an infinite loop..
11:05:10linuxstbmidkay: Try re-running ../tools/configure - or better still, delete your build directory and start again.
11:05:26midkaylinuxstb: i did each of those already, and then re-checked out.
11:05:40midkayno go
11:05:48linuxstbmidkay: Why does that path you quoted include ".local/share/Trash" ?
11:06:04midkayi have no idea.. just before the lines i pasted:
11:06:05midkay<imphasing> ADDCC r0,r0,#1
11:06:07***Saving seen data "./dancer.seen"
11:06:15midkayCC sysfont.c
11:06:15midkayAR+RANLIB /home/midkay/.local/share/Trash/files/rockbox/build/librockbox.a
11:06:39midkayrather weird, but.. not sure if it's supposed to happen or what
11:07:29linuxstbmidkay: Where (under /home/midkay/) is your "rockbox-devel" directory?
11:08:25linuxstb.. or "rockbox" if you just did "cvs co rockbox"
11:08:57midkaycvs co rockbox.. and, it's right in /midkay
11:09:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
11:09:10*imphasing is away: Sleeptime.
11:10:02linuxstbmidkay: Type "pwd" in your current directory - does it say "/home/midkay/rockbox" ?
11:10:23midkayum. no. weird.
11:10:28midkayoh. crap. wait.
11:10:36midkaywait. what happened.
11:10:52linuxstbI'm guessing you dragged and dropped the old rockbox folder to the trash, but didn't "cd" out of it from your terminal.
11:11:14midkaysomehow rockbox ended up in the trash.. well, i did delete rockbox-d.. aa..h...? :) hmm..
11:11:16midkaylet me try again
11:13:22midkayk, same error, but now in /home/midkay/rockbox-devel/ ...
11:14:19midkay/home/midkay/rockbox/build/apps/bitmaps/native/rockboxlogo112x37x1.c = "error - Unsupported compression 2
11:16:13linuxstbHave you changed that bitmap yourself?
11:17:00linuxstbbmp2rb works fine for me with that bmp file in CVS.
11:17:49 Join Zagor [0] (
11:18:03*LinusN spots a celebrity in the channel
11:19:01*Zagor waves
11:20:24 Quit darkless (Read error: 110 (Connection timed out))
11:21:06*LinusN is not worthy
11:21:22midkaylinuxstb: it's unchanged..
11:23:02B4gdermidkay: is this by any chance a 64bit host?
11:23:14midkayB4gder: yeah, it is :)
11:23:44B4gderthen the bad news is that bmp2rb doesn't work for you
11:24:12midkayi knew it'd be an awful idea to get the 64-bit version ffs.
11:24:45B4gdernah, its a golden chance for you to fix the rockbox build for 64bit!
11:24:47LinusNi think lots of things won't work for you
11:25:00midkayB4gder: that sounds fun, but... ;)
11:25:07midkayLinusN: uh-oh :\
11:25:24*safetydan wonders what preglow did to his EQ UI to stop it working
11:25:35safetydanit works here with 16.16 format q
11:26:36 Join mikearthur [0] (
11:27:41linuxstbsafetydan: I think preglow has broken the eq:;topicseen#msg17544
11:27:44Papricaarrr i give up.
11:27:57midkayLinusN or B4gder: any recommendations? can someone send me what i need to continue the build (some output files?) to see if anything else is broken?
11:29:16LinusNPaprica: give up what?
11:29:31Papricai try to load a bitmap for the icons
11:29:37linuxstbmidkay: You could test an older version of Rockbox - before bmp2rb was used in the build system. bmp2rb should be quite easy to fix if that's the only problem.
11:29:48Papricabut i without success
11:30:06LinusNPaprica: show me
11:30:13Papricablah what the"i" doing here
11:30:32Papricamm wait
11:30:33midkaylinuxstb: any suggestion on version? and, how? :) -d [date] or something iirc?
11:30:38Papricai try to change the code
11:30:58Papricabut it really dont work now
11:32:32linuxstbmidkay: bmp2rb was added to the build system on 22 January - try "cvs -d....etc... co -D 2006-01-22 rockbox"
11:32:45LinusNPaprica: can i see the src?
11:33:00linuxstbmidkay: that will give you the version at midnight on 22 Jan - before the change
11:33:40Papricasec... i want to see you the working one(loading the bmp and show it), and if you can help me to change it for work in the status bar
11:34:12midkaylinuxstb: k, checking out alright
11:35:15 Quit Xerion ("Leaving")
11:35:23Papricathis is working one,
11:35:37Papricabut when i try to change it to global (variables)
11:35:43Papricait doesnt work
11:36:00midkaypast the point i got the error before, still going..
11:36:17midkay.. error on using uclpack
11:36:18LinusNPaprica: show me the non-working code
11:36:36Papricablah, i dont have it =\
11:36:40Papricai try to do it again
11:36:47Papricaand let see what happend
11:41:08midkaycan i do anything to avoid uclpack being used at build?
11:42:35 Join einhirn [0] (
11:42:40linuxstbmidkay: I don't think so - I think uclpack is only used when it is needed.
11:43:14midkaylinuxstb: how do i make it unneeded? :)
11:43:21linuxstbMaybe Rockbox smaller...
11:43:33midkayit's only used for flashing, no?
11:43:59linuxstbNo, I think it's used when the normal build is too big to be loaded as a firmware upgrade by the Archos firmware.
11:45:21linuxstbYou could maybe try installing a different version of uclpack (it's a standard tool), and then edit the Makefile in your build directory to refer to that version, not the version in tools
11:46:22linuxstbAre you using Debian (or similar)?
11:46:34midkaysimilar? mandriva
11:46:52linuxstbUbuntu (for example) is based on Debian.
11:47:56midkayi'm using mandriva..
11:48:27linuxstbI don't know anything about mandriva. I also don't know very much about uclpack....
11:48:44linuxstbBut there's a package for Debian called "upx-ucl" which I think is the same thing.
11:49:03Papricadont work =\
11:49:09linuxstbMaybe mandriva has a package for ucl
11:49:38 Join tvelocity [0] (
11:50:20LinusNPaprica: what happens?
11:50:30midkaylinuxstb: i don't think it does, tried some different things and then looking for it
11:50:49midkayLinusN: can you confirm whether uclpack is run simply for rombox or is it actually _needed_?
11:51:15LinusNit is needed if the ajz file is too big
11:52:38midkaydamn.. hm.
11:52:48LinusNPaprica: i don't see you initializing the icon pointer in the statusbar struct
11:52:51 Join jack123 [0] (i=callme_j@
11:52:52midkayguess it's a reinstallation of mandriva 32-bit..
11:53:02LinusNPaprica: should it really be a pointer?
11:53:23jack123hello buddy
11:53:26jack123i am jack
11:53:31Papricadont know
11:53:34peturaargh.. bootloader stays in usb mode after disconnect, then after a while all text is gone :(
11:53:44LinusNPaprica: i don't see why it would need to be
11:53:57jack123is any girls from australia
11:54:06jack123who is paprica
11:54:08Mode"#rockbox +o LinusN " by ChanServ (ChanServ@services.)
11:54:26PapricaLinusN, ok
11:54:39LinusNjack123: this place isn't exactly crowded with girls
11:55:21jack123every server is busy in somthin in solve
11:55:31jack123i don't find any chat room................
11:55:36jack123to chat in frank
11:55:39LinusNso unless you aren't here to chat about Rockbox, i suggest you go somewhere else
11:55:40linuxstbmidkay: What error does uclpack give you?
11:56:04jack123what is rockbox
11:56:15midkayOBJCOPY rockbox.bin
11:56:15midkayUCLPACK rockbox
11:56:15midkaymake[1]: *** [/home/midkay/rockbox/build/rockbox.ucl] Error 1
11:56:15DBUGEnqueued KICK midkay
11:56:15midkaymake: *** [all] Error 2
11:56:17LinusNjack123: see Topic
11:56:23PapricaLinusN, works
11:56:28LinusNPaprica: wee
11:56:41jack123ok buddy
11:56:43linuxstbmidkay: Try "make V=1"
11:56:59 Part jack123
11:57:15Mode"#rockbox -o LinusN " by LinusN (
11:57:28midkaylinuxstb: same thing pretty much
11:57:29midkay(echo "UCLPACK rockbox" ; \
11:57:29midkay/home/midkay/rockbox/tools/uclpack −−best −−2e /home/midkay/rockbox/build/apps/rockbox.bin /home/midkay/rockbox/build/rockbox.ucl >/dev/null 2>&1;)
11:57:29midkayUCLPACK rockbox
11:57:29***Alert Mode level 1
11:57:29midkaymake[1]: *** [/home/midkay/rockbox/build/rockbox.ucl] Error 1
11:57:30***Alert Mode level 2
11:57:30midkaymake: *** [all] Error 2
11:58:17linuxstbTry running that uclpack manually, but remove the redirections at the end - i.e. type everything up to ".../rockbox.ucl"
11:58:28CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
11:58:28*petur just installed a backdrop image and is impressed
11:58:36 Join Rondom [0] (
11:59:34midkayucl_init() failed !!!
12:00:58Papricashit, LinusN, now when i try to view the image in the statusbar function, the simulator dont respond
12:01:47LinusNPaprica: hehe
12:02:03*LinusN wispers ...gdb....
12:04:21midkayi think i'll just go to 32-bit
12:05:04linuxstbmidkay: I think the place to look is in tools/ucl/include/ucl/uclconf.h - to fix it for 64-bit architectures. But I don't know immediately what is wrong.
12:05:12Papricawarning: Resume index 0 offset 6F8
12:05:13PapricaProgram received signal SIGSEGV, Segmentation fault.
12:05:13Paprica[Switching to thread 1648.0x1720]
12:05:13DBUGEnqueued KICK Paprica
12:05:13Paprica0x00000000 in ?? () from
12:05:52midkaywell, for one, i'll probably have more troubles like this if i don't go 32.. two, a lot of boring work to get stuff done if i need to "go around" everything like this.. and three, i'm not very set-up yet, better now than later.. only installed mandriva last night :)
12:06:46 Part petur
12:07:31***Alert Mode OFF
12:07:49midkaylinuxstb: haha, i've absolutely no idea what to look for :D
12:08:11 Join Moos [0] (
12:09:57linuxstbmidkay: No, I don't either.
12:11:41midkaylinuxstb: no problem, i'd rather just go 32-bit :) thanks a lot for all your troubleshooting though :)
12:14:11 Join Lost-ash [0] (
12:15:27 Quit midkay (Remote closed the connection)
12:16:21 Quit tvelocity ("Αναχωρεί")
12:16:46LinusNPaprica: you are probably still not setting your pointers correctly
12:17:07LinusNor are overflowing a buffer
12:17:22 Join midkay [0] (
12:17:26Papricaok will check it
12:23:31Papricalcd_bitmap works
12:23:40Papricadont work
12:25:59 Join NightCat [0] (
12:26:06NightCatHi all!
12:26:41NightCatSmall quest.: In bootloader v5 shoud Bootloader USB mode work?
12:27:27NightCatCause when I plugging in usb cable it boots up and writes "Bootloader USB mode", but then nothing happens...
12:27:50NightCatIt's about H300.
12:28:23stamppotNightCat: same here, I was gonna ask about that as well..
12:28:38NightCatAnd I want to talk with Paprica.
12:28:46NightCatPaprica here?
12:28:49 Quit ashridah (Nick collision from services.)
12:28:51 Nick Lost-ash is now known as ashridah (
12:29:13NightCatPaprica: Does you made margin patch?
12:29:39NightCatPaprica: I was looking on it and think about Y marging...
12:29:48NightCatWhat you think about this?
12:30:02Papricawhy do you need y margin?
12:30:22Papricayou get it from the line number*font height
12:30:30Papricayou talk about wps right?
12:30:32NightCatI was playing some around and make it do y marging only for static (not scrolling) text...
12:30:39NightCatPaprica: Yes.
12:31:15NightCatI need this cause with some fonts on the screen is for ex. 13,5 strings...
12:31:26NightCatAnd it starts from the top...
12:31:41NightCatAnd I have a half of a string at the bottom...
12:32:01NightCatI want to move strings to bottom.
12:32:16Papricaso didnt you ear about blank lines?
12:32:31NightCatI don't like this...
12:33:04NightCatIf I move it to bottom I could use 1,5 instead of 1 string for status pictures.
12:33:09LinusNNightCat: the bootloader usb mode is still shaky, it seems to depend on the "USB charge" setting in the iriver firmware
12:33:18NightCatAnd all will looks mach more beauty...
12:33:44NightCatLinusN: Ok. I'll try it now.
12:34:22Papricaso edit the patch
12:34:26NightCatPaprica: I could make a screenny to show it if you want?
12:34:49Papricagive me a screen dump..
12:36:13Papricathanks god, it works.
12:36:24NightCatWhat works?
12:36:48Papricathe start of custom statusbar icons
12:38:06PapricaLinusN, do you know why lcd_bitmap works fine and display->bitmap doesnt work?
12:38:40NightCatHere you are:
12:39:02NightCatYou see: there is some place at the bottom...
12:39:14LinusNPaprica: what do you mean with "doesn't work"?
12:39:23Bg3rNightCat what font do you use btw ?
12:39:46ender`is it just me, or is sourceforge down?
12:39:58PapricaLinusN, when i use display->bitmap the simulator doesnt respond
12:40:12linuxstbender`: Sourceforge is down.
12:40:20NightCatPaprica: What do you think?
12:40:29ender`ok, i thought it was my isp again
12:40:56Papricawithout the y margin
12:41:06ashridahsourceforge's been flakey all day :/
12:41:07NightCatPaprica: What?
12:41:15Papricayou couldnt do this wps?
12:41:36LinusNPaprica: what format is the bmp?
12:41:38NightCatThis WPS without y marging...
12:41:45Paprica24 bit
12:42:03NightCatBut I want to move all strings some to bottom...
12:42:09LinusNwhat format is the bmp for the remote?
12:42:32NightCatPaprica: I know, you can do this, please, try...
12:43:11NightCatPaprica: Third parameter for the %m will be good...
12:44:06linuxstbNightCat: You are talking a viewport?
12:44:06NightCatPaprica: for ex.: %m|2|218|5| - 5 pixels offset from the top of the i string...
12:44:30NightCatlinuxstb: We are talking about margings...
12:44:49NightCatlinuxstb: we have a patch witch sets X marging.
12:44:51Papricaso this is the problem?
12:44:54Papricathe remote?
12:44:59NightCatlinuxstb: And I want Y marging too...
12:45:16NightCatPaprica: What with the remote?
12:45:27PapricaNightCat, i talk with linus..
12:45:34NightCatOh sorry..
12:46:33PapricaNightCat, i will take a look on this in the evening
12:46:52NightCatPaprica: A big big thanks for you!!!
12:47:39NightCatPaprica: But when you'll looking don't forget to test on scrolling lines... Cause I have problems only there...
12:51:02LinusNPaprica: i don't know
12:51:14Papricaok i will check it.
12:51:20LinusNsinglestep in the debugger
12:52:50NightCatLinusN: USB connected picture could be 24 bit color?
12:53:36NightCatThen I can replace it with my new one?
12:53:38 Join actionshrimp [0] (
12:53:50NightCatOk. Wi'll try now...
12:54:48 Join petur [0] (
13:01:08peturanybody interested in reviewing my recording gain patch?
13:01:53Papricamm how can i know if the remote is drawing or the player?
13:02:32CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
13:02:32*B4gder admits to never ever having recorded anything with Rockbox ;-)
13:03:05*petur bought his H340 for recording (amongst others)
13:03:34LinusNPaprica: set a breakpoint in lcd_remote_bitmap
13:04:05PapricaLinusN, i know that the remote is the problem
13:04:35LinusNi just can't understand why read() doesn't read the entire codec file in the sdl sim version of codec_load_file()
13:04:59LinusNit only reads 8k of the 200k file
13:05:32Papricai only need to separate the bitmap and the mono_bitmap
13:05:41LinusNpetur: what's the deal with the two gain patches on sourceforge?
13:06:07LinusNPaprica: yes, you need to have two sets of icons, one for each lcd
13:06:11***Saving seen data "./dancer.seen"
13:06:29linuxstbLinusN: What is the errno that read() returns?
13:06:34Papricaok but how can i know which lcd is drawn?
13:06:36peturMmmm started tweaking my original patch and posted it as a new patch,
13:06:45LinusNlinuxstb: it doesn't return an error
13:07:00peturThe latest version is on my (original) entry...
13:07:34LinusNlinuxstb: it returns the number of bytes read (8366 bytes)
13:08:47linuxstbLinusN: I would replace it by my own read function that performs multiple calls to read up to the amount of data requested, or until an error.
13:09:25linuxstbBut I don't know why (or if) that would work.
13:13:01 Quit XavierGr ()
13:14:44LinusNlinuxstb: i tried that, but the remaining read() calls return 0
13:15:56linuxstbLinusN: Windows or Linux?
13:16:20 Join hshah [0] (
13:16:34 Quit DreamTactix291 (Read error: 110 (Connection timed out))
13:17:25linuxstbLinusN: If Windows, then I think O_BINARY could be the problem.
13:23:34peturB4gder: devcon will not be possible for me, agenda collision :( (damn, already had my wife almost convinced)
13:27:56 Part hshah ("Leaving")
13:35:01*[IDC]Dragon awakes
13:35:13[IDC]Dragonwhat's the devcon status?
13:35:29B4gdernothing new really
13:35:46B4gderwell, apart from petur not coming
13:36:03[IDC]DragonI don't even know the old
13:36:24B4gderthe only ones who's said they're coming is me, Linus and Zagor
13:36:37B4gderseveral people seems to be checking details and plans
13:36:51[IDC]Dragonhow surprising of you 3 ;-)
13:36:57B4gderyeah ;-)
13:37:01*[IDC]Dragon checks th m/l
13:37:12ashridahyeah, i can't be a groupie and fly over there, sadly.
13:37:24B4gderI guess we won't get a lot of problems with fitting us all in ;-)
13:38:07peturI could fly in saturday night and leave again sunday afternoon, but somehow it doesn't seem worth it...
13:38:18ashridahbesides, uni starts again in a few weeks, and i've got books+junk to buy
13:44:03[San]finally got doom to work
13:53:01hyariondoes rockbox have doom now?
13:53:31linuxstbA port is in progress, and appears to work on the iriver H300.
13:53:45hyarionoh, nice :)
13:54:02linuxstbAre you responsible for Doom on ipodlinux?
13:54:10hyarionyes I am :)
13:54:23linuxstbI've been wanting to ask - do you use the cop at all?
13:54:32linuxstb(in Doom)
13:54:33hyarionnope, not yet
13:55:40*hyarion is away
14:00:12LinusNlinuxstb: O_BINARY was the solution :-)
14:00:46 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
14:02:49linuxstbLinusN: Hurray!
14:03:03*linuxstb hates O_BINARY
14:03:14[IDC]DragonOT: does somebody have a hint for me on what to use to build a small server, with minimum idle consumption?
14:03:43[IDC]Dragonlike AMD Geode, Via, etc
14:03:57[IDC]Dragonbut with a big RAID
14:04:03 Quit [San] (Read error: 110 (Connection timed out))
14:04:04peturuse a pentium-M or AMD turion
14:04:16[IDC]Dragonand disks powered down when not accessed
14:04:19markunAnother guy using his nick instead of his real name in the wiki..
14:05:24linuxstbmarkun: Have you had much time to work on the gigabeat recently?
14:05:49[IDC]Dragonpetur: sounds like 30W already, I wonder if I can get away for less
14:05:58markunNo, I was very busy writing a paper. I made some changes to rockbox to get it to compile some more and started on the bootloader
14:06:46petur[IDC]Dragon: that would be at 100% CPU load...
14:06:52linuxstbmarkun: Which arm architecture is your processor again? Does it have more instructions than the arm7tdmi?
14:07:14[IDC]Dragonpetur: really? well then
14:07:26[IDC]Dragonit depends on the chipset, too
14:07:31markunIt's an arm9tdmi, also ARMv4
14:07:54[IDC]Dragonand I wonder if Linux supports something like disk powerdown
14:08:48peturit does support scaling back cpu freq and voltage based on load (I think)
14:09:24linuxstbmarkun: Any idea what the differences are? I'm just wondering if CONFIG_ARM is too general, or if it's going to work for the gigabeat.
14:10:03B4gderI _think_ its gonna work as is
14:11:04markunOne of the differences is that it has a MMU. The instructions are all the same I think.
14:12:56 Join Rondom [0] (
14:13:51 Join elinenbe_ [0] (
14:13:56 Quit elinenbe (Read error: 104 (Connection reset by peer))
14:14:05 Nick elinenbe_ is now known as elinenbe (
14:14:06B4gder"equivalent to a 27% performance increase" at the same frequency
14:14:13 Join muesli- [0] (n=muesli_t@
14:14:24B4gdersimply faster/better pipeline
14:15:40 Join elinenbe_ [0] (
14:15:41 Quit elinenbe (Read error: 104 (Connection reset by peer))
14:15:50 Nick elinenbe_ is now known as elinenbe (
14:19:08 Part LinusN
14:31:00 Join Matze41 [0] (
14:35:31MoosHi muesli !
14:36:33*amiconn appears
14:36:46MoosHello to amiconn too
14:38:10amiconnB4gder: As you often mention it ('time to say moo'), do you mean this one: ?
14:39:22B4gderbut really, its just a silly phrase with no meaning
14:39:39amiconnAh :)
14:46:01 Quit mikearthur (Remote closed the connection)
14:46:30 Join mikearthur [0] (
14:47:25 Quit Matze41 ("Miranda IM! Smaller, Faster, Easier.")
14:48:02 Join Matze41 [0] (
14:48:49 Quit Matze41 (Client Quit)
14:49:10 Join Matze41 [0] (
14:53:44 Join damaki_ [0] (
14:55:36linuxstbamiconn: Is this what you had in mind for the pixelfuncs?
14:55:36 Join DreamTactix291 [0] (
15:03:50 Quit B4gder ("time to say moo")
15:06:15***Saving seen data "./dancer.seen"
15:07:43 Quit damaki (Read error: 110 (Connection timed out))
15:13:43 Join Kohlrabi [0] (
15:17:12 Quit DreamTactix291 (Read error: 110 (Connection timed out))
15:18:59 Quit ashridah ("Leaving")
15:20:15 Quit jaebird (Read error: 113 (No route to host))
15:20:25 Join LinusN [0] (
15:21:40PapricaLinusN, the statusbar icons load from a file.
15:21:53 Quit muesli- ("ich will Khe!!!")
15:22:08LinusNPaprica: nice
15:24:38 Join [San] [0] (
15:25:51[San]how do I get the backdropp working? Is it more than just putting the image in /rockbox/backdrops/ ?
15:27:19Paul_The_Nerd[San]: You have to browse to it, open the context menu (long hold "Select" or your target specific equivalent) and choose "Set as Backdrop"
15:27:54Paul_The_NerdAnd if it's in /.rockbox/backdrops/ it will be saved across reboots etc.
15:28:30[San]and I see bmp viewing is working now
15:30:55[San]ah, it works
15:30:59[San]thanks Paul
15:31:04Paul_The_NerdNo problem
15:32:24 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
15:33:20 Quit Matze41 (Read error: 104 (Connection reset by peer))
15:35:37[San]I love this one
15:37:59Papricacustom icons on statusbar =]
15:38:14[San]new patch?
15:38:38Papricamm im working on that..
15:39:05Paprica - the icons file..
15:40:32 Join jaebird [0] (n=jaebird@
15:40:49MoosPaprica: that sounds really promising
15:41:14[San]i need the background patch to be included with the optimised build
15:41:20[San]then I can change the text color
15:42:11Papricawe need to do a "mark picture" also
15:44:41Paul_The_NerdMark picture?
15:44:45 Quit merbanan (Read error: 104 (Connection reset by peer))
15:45:05Papricai dont found another name
15:45:48Paul_The_NerdYou mean the line selector?
15:45:50Papricawhen you over text...
15:45:58 Join Matze41 [0] (
15:46:32Paul_The_NerdWith skinning of that, and a viewport setting, there will be significantly less room for criticism from the people who say "rockbox is ugly"
15:49:54[San]i have an idea
15:50:22[San]is there code in the source that defines how far across the names in the file browser are ment to be?
15:50:25 Join linuxstb_ [0] (
15:50:39 Join webguest63 [0] (
15:51:59linuxstb_Paprica: How do you want to customise the line selector? Do you just mean change the colour?
15:52:13 Quit linuxstb (Nick collision from services.)
15:52:14 Nick linuxstb_ is now known as linuxstb (
15:53:13 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
15:53:29linuxstb:) You may have a hard time persuading people that that is needed.
15:53:38 Join Kohlrabi [0] (
15:53:50Papricalol =]
15:57:04Paul_The_Nerdlinuxstb: At the very lease, gradients would looks really nice. :-P
15:58:54linuxstbWe already have problems with LCD speed on the H300...
16:01:09 Quit Zagor ("Client exiting")
16:02:28webguest63Paprica: your scroll margins patch doesn't work annymore :(
16:02:56Paul_The_NerdMan, "would looks really nice?" Apparently my grammar's about as bad as the idea. ;-)
16:03:37linuxstbPaul_The_Nerd: Have you had problems with the buttons on your Nano in the last 24 hours or so? (after I committed the change to remove the udelays from the button driver)
16:04:12Paul_The_Nerdlinuxstb: Not at all, and I've been running with a version after that commit.
16:04:47linuxstbStrange. I'm sometimes getting "button freezes", but a few scrolls on the wheel brings it back to life.
16:05:05Paul_The_NerdI haven't had any problems here.
16:05:34linuxstbAnd no-one has complained on the forum or here...
16:06:06Papricain the simulator it worls
16:06:07Paul_The_NerdWell, we don't know how many people are updating often enough on 4g targets to stumble across it yet
16:06:12Papricain the player, not
16:06:22Paul_The_NerdRemember, the Nano clearly handles things at least a bit differently, since it works without the udelays at all.
16:07:34linuxstbYes, but preglow reported some problems as well.
16:07:54linuxstbMaybe we've just worn out our wheels...
16:07:59Paul_The_NerdWith the other udelay still removed or with it set how it is in CVS?
16:08:12linuxstbNot sure - probably with them both removed.
16:08:27Paul_The_NerdSo, maybe it's a symptom of it being too low. 10 for 4gs, 0 for Nanos?
16:08:58linuxstbI'm going to try that now, see if it fixes my occasional problems.
16:09:00Paul_The_NerdI've got it as CVS, I only tried it with both removed briefly, but saw no reason to leave it different from CVS in practical terms.
16:09:12[San]is there code in the source that defines how far across the names in the file browser are ment to be?
16:09:46 Part petur
16:11:14[San]or can be?
16:14:05linuxstb[San]: AFAIK, all text is drawn the full width of the screen at the moment - there is no option to specify a width.
16:14:05LinusN[San]: that is defined by the font
16:14:25[San]well, not that
16:14:36LinusNah, i thought you meant the line spacing
16:14:42[San]kind of
16:14:47 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
16:15:04[San]like, how far out the names of the folders of is out from the side of the screen
16:15:11[San]*folders is
16:15:22[San]*folders are
16:15:48[San](sorry, i'm still half asleep)
16:16:27 Join DangerousDan [0] (
16:16:29Paul_The_NerdSo, horizontal margins essentially? Draw text between starting at pixel X1 and ending X2 only, scrolling if wider than that?
16:16:41LinusNsafetydan: u there?
16:16:56[San]like, if I wanted to move all the folder names out 40px from the edge..
16:17:21linuxstb[San]: Are you talking about only the folder names, or both the folder names and filenames?
16:17:36[San]like, all, even settings and stuff
16:18:16linuxstbNo, but hopefully the talked-about viewport facility will enable that to happen.
16:18:24[San]ah, ok
16:18:26kernelsenseihow to define the background color in a h300 wps ?
16:18:53lostlogicwhat do we think of a "backlight->hold" option where the backlight follows the hold switch (as mentioned on the forums)?
16:19:41Paul_The_Nerdkernelsensei: Create a bitmap of the color(s) you want, that is the size of the screen, and load it with %X|filename.bmp| (capital X)
16:19:52*LinusN makes a huge SDL commit
16:19:57*Paul_The_Nerd cheers.
16:20:09Mooswee !
16:20:31LinusNlet's hope i didn't forget anything
16:20:38kernelsenseiPaul_The_Nerd: ok thanks
16:20:59*Paul_The_Nerd looks forward to the upcoming "oops" then.
16:21:07LinusNPaul_The_Nerd: inevitable
16:21:22Paul_The_NerdBut, it just wouldn't be the same without one.
16:21:29linuxstbAnyone taking bets on Linus's score?
16:24:02 Quit Rob2222 (Read error: 104 (Connection reset by peer))
16:24:09 Join DreamTactix291 [0] (
16:24:12PapricaLinusN, can you take a look on the good and guess why it doesnt work on the player?
16:24:36Papricain the simulator it works..
16:24:37LinusNthe archos player?
16:24:54Papricaon h300..
16:25:01Papricain the simulator it eorks..
16:25:12LinusNwhat happens on the player?
16:25:15Papricabut in the player doesnt
16:25:30Papricadont view the pics.
16:26:15 Join Xerion [0] (
16:27:20LinusNwhere does it load the bmp?
16:27:29safetydanLinusN: yo
16:27:40amiconnlinuxstb: Yes, that's what I had in mind (you can keep the 2 arrays const, btw), plus an obvious optimisation for lcd_clear_display()
16:27:46LinusNi just committed your sdl patch
16:27:54safetydanoh cool
16:28:14PapricaLinusN, you mean what is the function?
16:28:28safetydanI know why the codecs don't load on Win32 too, but I can't fix it since it involves Makefile magic
16:28:53Papricalin 64..
16:28:56LinusNPaprica: ah, found it
16:29:15linuxstbamiconn: The compiler gave me warnings if I kept the arrays as const. Which is why I wasn't sure I should commit it.
16:29:20LinusNPaprica: i don't see any obvious mistakes
16:29:35LinusNsafetydan: i fixed that
16:29:51LinusNsafetydan: without makefile magic
16:29:54Papricathis patch got me crazy
16:31:37safetydaninteresting. I guess I'll find out how you did that when the diff shows up in the log
16:31:47 Join Rob2222 [0] (
16:32:30LinusNsafetydan: the key was the O_BINARY option to open()
16:32:47LinusNsilly windows
16:33:04amiconnlinuxstb: 'const' with pointers is sometimes a bit tricky, but it should still be possible
16:33:11amiconnI'll try that tonight
16:33:50amiconnThe compiler distinguishes whether the pointer is constant, the data pointed to is constant, or both
16:34:05safetydanwell darn, I would never have thought to check that
16:34:27linuxstbamiconn: OK, thanks. Happy to leave it to you.
16:35:16amiconnLinusN: Imho O_BINARY isn't silly
16:36:02LinusNamiconn: text mode file access is silly imho
16:36:32Papricathe 4 folders of the Boxes wps
16:36:39Papricavery angry
16:38:17lostlogicoops, got distracted by real work. Backlight->Hold would be an option in addition to Off, On and 1s...90s where the backlight would be on exactly when the hold switch is not.
16:40:38linuxstbWell, I don't think it's needed on the ipod - just touching the scrollwheel brings the backlight back on, and a tap doesn't register as a button event.
16:41:59PapricaLinusN, so you dont know why it doest work on the player?
16:42:08amiconnLinusN: Maybe, but then this differentiation is definitely not windows specific
16:44:57 Quit Matze41 ("Miranda IM! Smaller, Faster, Easier.")
16:46:44LinusNamiconn: maybe not windows specific, but still very silly
16:46:49lostlogiclinuxstb: so you can adjust volume on iPod w/o backlight going on already?
16:47:10[San]is DAP review down?
16:49:14Paul_The_Nerdlostlogic: So, the hold switch would lose its hold functionality in favor of not turning the BL on?
16:49:32Paul_The_NerdTHAT is a SCORE
16:50:29linuxstblostlogic: No, "moving" the scrollwheel changing the volume (and hence activates the backlight), but touching it just activates the backlight. (in the WPS)
16:50:57linuxstbPaul_The_Nerd: hehe. I would never have gone that high...
16:51:08[San]dap review is down
16:52:05DBUGEnqueued KICK [San]
16:52:05*[San] Skinny Dips in a sea of Dream Theater - A Sort of Homecoming - Nyc Radio Ad
16:52:09[San]ever hear that?
16:52:18Paul_The_NerdBuuuuuut... the H120 sim is fine.
16:53:03[San]crap wrong window
16:53:53NightCatWhy there is no option "Sort current playlist"?
16:53:56lostlogicPaul_The_Nerd: no, it would also lock the buttons, but the backlight would just follow it... the theory being that some people will only need to see the screen when the unit's buttons can be activated.
16:54:00linuxstbPaul_The_Nerd: I think the h120 sim is still the x11 build
16:54:47 Join NicoFR [0] (
16:54:55lostlogiclinuxstb: the point isn't being able to activate the backlight w/o a button press, the point is to be able to flick it on for exactly as long as you need to either read the playing track, or change tracks, and flick it back off with no delay.
16:55:19Paul_The_Nerdlinuxstb: Ah, yes, it would be that they're all color-screened.
16:55:58linuxstblostlogic: OK. But I have a short timeout on my backlight now - simply because it is very easy to reactivate it just by touching the wheel.
16:56:40linuxstbI'm not saying it's not useful - I'm just saying I don't need it.
16:57:08lostlogic*nod* yeah, would be less useful on iPod than perhaps other devices because of the touchiness
16:57:27Paul_The_NerdI can see it being nice.
16:57:46linuxstbIt's a hard option to describe in two or three words though.
16:58:15 Quit Xerion (Remote closed the connection)
16:58:23Paul_The_Nerd"Off on hold"?
16:58:37linuxstb(it took lostlogic two lines...)
16:59:44linuxstb"Off on hold" sounds good.
17:00:00LinusN4654 is definitely not a bad score on the daily build page!!! :-)
17:00:42[San]misticriver is down now
17:00:42linuxstbYou're lucky there are only 4 SDL sims being built....
17:01:03mikearthurwhat are the scores?
17:01:07mikearthurhow many hits?
17:01:46LinusNPaul_The_Nerd: there came the "oops" :-)
17:02:00Paul_The_NerdLinusN: You have exceeded all expectations. I congratulate you.
17:03:53 Quit NightCat ()
17:06:16linuxstblostlogic: Has anyone tried your "null output" patch on the ipod?
17:06:20***Saving seen data "./dancer.seen"
17:06:39lostlogiclinuxstb: didn't preglow? it should work on any swcodec the same
17:07:07linuxstbI can't remember. I want to try it tonight - see exactly how the various codecs are running.
17:07:37linuxstbI also want to try and get alac running better. It shouldn't be the slowest codec...
17:08:23lostlogichehe, I think I'm going to try and hit up that developers conference afterall...
17:08:28Paul_The_NerdDoes ALAC slow down the UI as horribly as MP3, or does it just have playback skips and such?
17:09:12lostlogiclinuxstb: can also profile them to help with the optimization once you know which are slow :)
17:09:21linuxstbPaul_The_Nerd: Both.
17:10:01linuxstblostlogic: Is the profiling code in CVS now?
17:11:25lostlogiclinuxstb: yeah
17:11:46lostlogiclinuxstb: there's doc in the tech doc file on how to turn it on and uhh stuff
17:13:35linuxstblostlogic: Found it now. I thought I remember you writing something, but I was searching the wiki for it...
17:14:50linuxstblostlogic: alac is going to be fun to profile - most of the functions are declared inline...
17:15:38Paul_The_Nerdlostlogic: "uhh stuff?" ;-) Such descriptive terms.
17:16:18lostlogiclinuxstb: the inlines will still be profiled unless you mark them with the NO_PROF_ATTR.. but that will mean that without that attr on them it will run SLOW AS ALL GETOUT
17:19:35linuxstblostlogic: So if I do nothing, and leave the functions as "inline", then are you saying gcc will not inline them when I enable profiling?
17:25:35 Join needleboy [0] (
17:25:42needleboygood evening everyone
17:26:00needleboycould anyone please take a look here, see if they could help? thank you!
17:26:08elinenbegood evening needleboy
17:26:24needleboyhey elinenbe
17:27:58[San]ooh, thats a big problem
17:28:00Paul_The_NerdNeedleboy: The color changer and the scroll margins were both broken by the background image patch. As it is, it's the patch creator's duty to maintain or update patches that don't work with CVS, not the other way around.
17:28:05needleboyyes it is....
17:28:14needleboyi know...
17:28:19 Nick Lynx_ is now known as Lynx_awy (
17:28:20Paul_The_NerdAt least I *think* they were broken by the background image one.
17:28:33[IDC]Dragonlinuxstb, r u there?
17:28:34needleboyi'm just asking for help if anyone can... the patches are Paprica's and he's not around ATM...
17:28:50Paul_The_NerdThere's actually work being done on a more flexible text color changer by JdGordon if I recall.
17:28:53needleboy2 of the patches are Paprica's....
17:29:03[San]needleboy, they can now set a background color in a WPS with an image
17:29:09needleboyhe ben :)
17:29:23needleboySan, but not the text color...
17:29:36needleboyben, i've ICQed you just now
17:29:43needleboythough you weren't there...
17:29:48Papricawhy 2 of the patchs are mine?
17:29:56CtcpIgnored 2 channel CTCP requests in 0 seconds at the last flood
17:29:56*safetydan looks at page
17:29:58safetydanI'm famous!
17:30:07needleboythink you can get yours fixed? :p
17:30:28needleboyoh right...
17:30:32[San]if we can get the text color fixed, we'er good then, right?
17:30:33Papricammm but why 2?
17:30:36needleboythe color changer is Pi's...
17:30:40linuxstb[IDC]Dragon: Yes.
17:30:41needleboyonly one :)
17:31:03Papricamm i think i will add a tag for the wps
17:31:08Papricathat change the color
17:31:14[San]I was thinking that
17:31:30needleboybut the menus will still be the default black.......
17:31:35[IDC]Dragonlinuxstb, I found another CLZ:
17:31:37needleboyso it doesn't help a lot...
17:31:40Papricaoh right =]
17:31:43needleboywe still need a color changer.
17:32:11linuxstb[IDC]Dragon: Thanks, but that one is already in my collection... :)
17:32:14[San]or we dont need it in the menu anymore, just in the wps
17:32:15Papricain my opinion the best color for the menus is black and white
17:32:23[San]or the whole player setting
17:32:24 Join Jolt [0] (
17:32:26Paul_The_Nerdpaprica: <−−- JdGordon made this, but it's not connected anywhere yet.
17:32:29Papricablack text, white background
17:32:47[IDC]Dragonthe new one in cvs doesn't look too good either
17:33:09linuxstb[IDC]Dragon: No, I have some better ones - but I'm trying to decide which one to commit.
17:33:20[San]thats only a plugin though?
17:33:21PapricaPaul_The_Nerd, i will give him the honor to complete this patch ;]
17:33:45Paul_The_NerdLinusN: Down to 121
17:34:02 Join damaki [0] (
17:35:01 Join muesli__ [0] (n=muesli_t@
17:37:38lostlogiclinuxstb: yeah
17:37:45linuxstb[IDC]Dragon: I'll probably commit the version you've just posted as the default C implementation, and I'm trying to do an ARM version. It would be so much easier if CLZ was there...
17:38:42[IDC]Dragonwhat's it used for, btw?
17:39:33 Part safetydan ("Leaving")
17:39:39linuxstbto count leading zeros... :)
17:40:01[IDC]Dragonin context, I mean
17:40:21linuxstbI don't understand the ALAC codec that well.
17:40:41 Join DJ_TCK [0] (
17:40:55[IDC]Dragonone of the few uses I know is float renorm
17:42:52linuxstbYou can look in apps/codecs/libalac/alac.c if you're interested. I don't claim to understand the maths behind lossless compression.
17:43:36linuxstb(it's called twice in that file)
17:43:52 Quit needleboy ()
17:45:56[IDC]DragonI found more:
17:47:39linuxstbThanks. I'll give that a go now.
17:47:39[IDC]Dragonand an faster one with a 256 enty table:
17:47:51 Quit damaki_ (Read error: 110 (Connection timed out))
17:48:03[IDC]Dragon(feel swamped)
17:50:43[IDC]Dragonsorry, I was too quick, the code does some other trick
17:50:51[IDC]Dragonbit reversal or so
17:51:12[IDC]Dragonbut something where CLZ is useful for
17:53:13 Join xmixahlx [0] (n=xmixahlx@
17:53:28[IDC]Dragonthe posting is only in google cache
17:54:51[IDC]Dragonlink to google cache:
17:57:42linuxstbThanks. Some more things to test :)
17:58:37 Quit hardeep ("Read error: 1 + 5 = --5 (Connection reset by third grade math)")
17:59:52JoltQuestion: Is verging on copyright infringement, as they are based on previous Winamp skins, hence they may not be allowed be included in CVS. Or are they edited enough to claim to be their own
18:00:15*[IDC]Dragon waves
18:00:38 Part [IDC]Dragon
18:04:02 Quit youngcereal ("Verlassend")
18:06:59linuxstbJolt: Are you the author of those WPSs?
18:07:02Paul_The_Nerdlinuxstb: So, I'm having wheel problems now on my Nano. But *only* after using the hold switch.
18:07:16linuxstbPaul_The_Nerd: Then don't use the hold switch :)
18:07:39Paul_The_NerdClearly an interim solution, but I thought it might interest you, since it didn't have these problems before the udelay changes. :)
18:08:04linuxstbWhat kind of problems? The ones I described?
18:08:41Paul_The_NerdBasically, if I switch hold on then off again, none of the buttons are responsive. I'm not wholly sure if it's for a certain amount of time, or until I just fiddle with them enough.
18:08:49 Join Thus0 [0] (
18:09:27linuxstbPaul_The_Nerd: OK. Could you try increasing those delays a little (I've currently set them to 10 and 10 and it seems OK) ?
18:10:38 Quit Thus0 (Client Quit)
18:14:46Joltlinuxstb: Yes
18:15:52 Quit DreamTactix291 (Read error: 110 (Connection timed out))
18:16:50Paul_The_Nerd10/10 at least still suffers from this.
18:18:20*Paul_The_Nerd starts experimenting with values
18:22:01linuxstbJolt: Did you create those images yourself, or did you take existing bitmaps created by someone else?
18:22:34Paul_The_Nerdlinuxstb: 20/20 seems to work.
18:23:20 Quit StrathAFK (Read error: 104 (Connection reset by peer))
18:23:28linuxstbPaul_The_Nerd: Have you tried 20/0 ?
18:23:36Paul_The_NerdI'll do that now.
18:23:58Paul_The_Nerd20 in read, 0 in int?
18:25:12linuxstbJust change whatever is 10 in CVS to 20 - I've forgotten which way they were.
18:26:33Paul_The_NerdApparently, CVS was opposite of what I said.
18:27:41 Join Strath [0] (
18:27:44Paul_The_NerdNeither combination of 20 and 0 works
18:28:09Joltlinuxstb: Took existing bitmaps created by someone else...........
18:29:30 Quit elinenbe (Read error: 104 (Connection reset by peer))
18:29:42 Join elinenbe [0] (
18:29:47 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
18:30:25Paul_The_NerdLinuxstb: Adding 10 to both (20/10) doesn't work. It seems 20/20 is a minimum for the hold switch to function properly
18:36:18 Join bluey [0] (
18:38:12 Join jack123 [0] (i=callme_j@
18:40:18 Part jack123
18:41:04 Join Febs [0] (
18:41:12 Quit imphasing (Remote closed the connection)
18:44:48 Quit DangerousDan (Read error: 104 (Connection reset by peer))
18:45:12*Jolt Jolt|Away
18:45:44 Nick Jolt is now known as Jolt|Away (
18:45:50 Join imphasing [0] (
18:46:05imphasingFor some reason, I don't have a bootloader directory in my tree..
18:47:30linuxstbJust type "cvs co bootloader" (from the directory containing apps, tools, firmware etc)
18:47:53imphasingah, cool
18:48:18linuxstbYou probably did "cvs co rockbox" instead of "cvs co rockbox-devel"
18:48:25 Quit Febs ("CGI:IRC (EOF)")
18:49:33 Join darkless [0] (
18:51:54 Quit [San] (Read error: 110 (Connection timed out))
18:56:31 Join jjoao [0] (n=joao@
18:57:03jjoaoHi for all
18:57:59linuxstbPaul_The_Nerd: Still happy with 20/20 ? If so, I'll commit it to CVS.
19:01:09 Join DangerousDan [0] (
19:01:35Paul_The_Nerdlinuxstb: As far as I can tell, 20/20 is fine. Haven't come across any additional weird behaviour.
19:01:47jjoaolinuxstb: Yesterday I was searching for DAP to buy. I found ipod Photo 60gb and iriver H120. Which one is the best to rockbox, My targets are only listen to music, great battery life and LCD with small fonts and a lot information (means great WPS)?
19:05:36 Join mafj2 [0] (n=mafj2@tor/session/x-7d66d6481b51808e)
19:06:22***Saving seen data "./dancer.seen"
19:07:36 Join tucoz [0] (
19:07:38 Join Philip_0729 [0] (
19:07:46 Quit Paul_The_Nerd ("Leaving.")
19:08:33 Part tucoz
19:10:04 Part Philip_0729
19:25:50 Join Therx [0] (
19:27:00 Quit Therx (Client Quit)
19:27:27 Join damaki_ [0] (
19:30:14 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:34:34 Join Sinbios [0] (
19:36:29 Join safetydan [0] (
19:42:34safetydanright, time to implement a slider based EQ UI
19:42:57 Nick Sinbios is now known as Medion (
19:43:02 Quit damaki (Read error: 110 (Connection timed out))
19:43:11 Nick Medion is now known as Sinbios (
19:46:02muesli__jjoao the h120: battery life is great, rockbox runs perfect on that one
19:54:40 Join damaki__ [0] (
19:55:08 Quit bluey ("Leaving")
19:55:20 Quit xmixahlx ("blah blah blah")
19:55:31 Join DreamTactix291 [0] (
19:56:07 Join Thus0 [0] (
20:01:46jjoaomuesli: how about the WPS in Ipod photo? Is it possible the full customization? I would like a lot of information on LCD and use use small fonts
20:02:13 Join einhirn [0] (
20:03:13stamppotjjoao: The WPS-es that you can use are the same across the different rockbox targets.
20:03:49 Join Therx [0] (
20:03:55stamppotThe only difference is that you can use color pictures only on color targets (but you probably figured that).
20:04:02jjoaostamppot: i see
20:04:45 Join linuxstb_ [0] (
20:05:26stamppotThe differences are more in the hardware and the physical resolution of the LCD
20:06:09jjoaoWhich one can show more information? forget the color
20:06:14stamppotFor instance, the H120 has a resolution of 176xsomething whereas the iRiver H300 has 226x176 as resolution
20:06:26linuxstb_jjoao: I would agree that the h120 is a better choice, mainly due to battery life. I'm not sure if Rockbox on the ipod will reach the battery life of the h1x0 port.
20:07:35linuxstb_jjoao: The H120's LCD is 160x128, the Photo is 220x176. But Rockbox has very small fonts - which means you can fit everything you want on the h120.
20:08:06 Join Chamois [0] (
20:08:14jjoaolinsuxstb: Yes, I loved the WPS gallery
20:08:49 Quit Chamois (Client Quit)
20:09:56jjoaoI am reading the IRC history to help me to choose a DAP, but I am still confuse. There is no doubts that H340 is the choice, but out of stock. I am findind H120, but is twice the price of ipod 60gb
20:10:36Sinbioshow is the toshiba gigabeat?
20:10:38linuxstb_Do you need optical I/O, recording or FM radio?
20:10:47Sinbiosi got one, but the software is crappy as hell
20:10:56Sinbiosso i'm anticipating a rockbox release :3
20:11:25jjoaoSinbios: I am sure that rockbox is the best GUI, I forgot the rest
20:12:05linuxstb_jjoao: In which case, the main difference will be battery life. Early tests show that Rockbox (without any optimisation) lasts about 7h-8hr, but Rockbox on the h120 can probably do twice that (I'm sure others will correct me).
20:12:06Sinbiosits hardware capabilities are pretty good, i think
20:12:21linuxstb_(7h-8h on the ipod)
20:12:59 Quit damaki_ (Read error: 110 (Connection timed out))
20:13:51linuxstb_Sinbios: The gigabeat port is still in the _very_ early stages - no code is running on the gigabeat yet.
20:14:15Sinbiosso i am a sad puppy :(
20:14:29Sinbiosi'm going to see if toshiba fixes the v3 firmware
20:14:32jjoaois the not complete tasks on ipodport a problem to brownse, mannagement and listen to music on ipod?
20:17:26 Quit DreamTactix291 (Read error: 110 (Connection timed out))
20:17:32 Join damaki_ [0] (
20:17:51linuxstb_Rockbox is usable today on the ipod - I happily use it to listen to music. There are a few bugs in the audio code (skipping tracks stop audio, but seeking a tiny bit in the file revives it), and the main missing feature is the battery status indicator.
20:18:29linuxstb_Most of the work left to do is optimisation - working on increasing battery life, and making the codecs run faster.
20:18:42Sinbioswow, the gigabeat has a 300mhz proc o_O
20:18:46Sinbiosvs. the ipod's 75
20:19:21linuxstb_And it's a better ARM than the ipod as well - so it is in effect more than 4 times as fast (but the ipod has two 75MHz arms)
20:19:46jjoaois the H340 rockbox finished too?
20:20:17Sinbioswell, if the ipod can manage video with two 75's, the gigabeat should be fine, right?
20:20:36linuxstb_Rockbox is never finished. The H340 port is almost at the same stage as the H1x0, but power management isn't complete, and the LCD driver needs a lot of work to speed it up.
20:20:54Sinbiosheck, even my ancient linux box can handle video with 300mhz
20:21:05linuxstb_Sinbios: The ipod doesn't manage video with two 75Mhz - the iPod Video has a video DSP that does the video decoding.
20:21:18linuxstb_and ipodlinux plays uncompressed video.
20:21:29Sinbioswait, uncompressed o_O
20:21:54Sinbioshow does it manage to fit any length of uncompressed video?
20:22:24linuxstb_It doesn't :)
20:23:01linuxstb_But I'm sure compressed video of some kind will eventually happen on the "normal" ipods.
20:24:56NicoFRlinuxstb: do you know if anyone is currently working on optimization for the h300 lc driver ?
20:27:22solexx_wow, new highscore on the daily builds page! Congratulations, Linus! :)
20:30:54amiconnlinuxstb: Rockbox does have less runtime than stock firmware on H1x0 as well (but more than on H3x0)
20:31:25jjoaolinuxstb: as I was say, if I dont need optical I/O, recording or FM radio, would be the ipod 60gb the best choice for me? is it possible that ipodport finish the development without solve the few bugs that it have?
20:31:44 Join Lear [0] (
20:31:46*amiconn wonders what's the deal with alac on iPod
20:32:10amiconnlinuxstb: Aren't most codecs actually faster on iPod than on iriver?
20:32:27Learpreglow: just tried the equalizer. didn't sound that good...
20:32:43Learsome sort of high frequency "pings" ruin the effect...
20:32:59amiconnLinusN: Congratulations to your top score (4654) ;)
20:33:10 Quit damaki__ (Read error: 110 (Connection timed out))
20:35:01linuxstb_amiconn: Yes, alac is a mystery.
20:35:31 Join gtkspert_laptop [0] (
20:35:40amiconnIt's ironic - apple's lossless format gives problems on apple hardware...
20:36:06midkayamiconn, huh?
20:37:07amiconnlinuxstb: Does lostlogic's profiling patch work on iPod?
20:38:06 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
20:39:06 Join Rondom [0] (
20:39:09amiconnBagder: Is there a build score top-ten somewhere?
20:39:17 Join Bger [0] (n=Bager@
20:39:40ghode|afk how are the scores calculated?
20:39:48 Quit Rondom (Read error: 104 (Connection reset by peer))
20:40:49solexx_ghode|afk: look at it, think about it and you'll find out :9
20:41:37solexx_(10 points for an error, 1 for a warning)
20:42:09 Join petur [0] (
20:42:33*safetydan feels slightly bad for LinusN as it was his patch that got him the high score
20:43:22Bgersafetydan don't worry, LinusN isn't that kind of man;)
20:43:38lostlogicI'm just happy that I am no longer the high scorer since the scores were created :-P
20:44:18 Join _FireFly_ [0] (
20:45:28 Quit elinenbe (Read error: 110 (Connection timed out))
20:52:57Learpreglow: don't worry too much; I might have found the reason... :)
20:53:17Learsafetydan: you did the eq ui, yes?
20:53:35 Join Rondom [0] (
20:53:36safetydanWorking on a new version now
20:53:58Learfound a small bug in your v5 patch: Order mismatch of settings between dsp.c and settings.h...
20:54:38Leardidn't sound very well the way it was... :)
20:54:52safetydanI'm not surprised
20:55:20hardeepdoes the sdl sim work on cygwin?
20:55:48safetydanshould do
20:55:54Learit should, as SDL is ported to Cygwin. Haven't tried to install it yet though...
20:55:58safetydanThat's what I tested my patch in
20:56:30 Join Febs [0] (
20:56:37hardeepcool, thanks
20:56:54safetydanLear: now I see... don't know how I missed that
20:57:49LearSeems to work much better now.
20:59:35LearHm.. You can browse presets, but not create them?
20:59:35Bgerlostlogic ?
21:00:18safetydanLear: they're stored in the config, you can use Manage Settings to save them out
21:01:04jjoaoI found this video in the internet, I think that is ipod Photo 40gb Rockbox.
21:01:07LearYes, but nothing to write just them, and to the right place?
21:01:30lostlogicBger: ?
21:01:38safetydanno not yet
21:02:05Bgerlostlogic i have q regarding charging status
21:02:15Bgerhow can i get this info?
21:02:22Bger(whether the battery is charging or not)
21:02:56Bgeri'm asking you because u did some work for h300 in this area recently
21:03:09lostlogicyeah, are you asking how to find it on the H3x0 or in general?
21:03:22Bgerin general
21:03:32 Quit Rondom (Read error: 113 (No route to host))
21:03:44Bgerbtw ... is it working on h300?
21:04:05lostlogicyeah, it works fine on the H3x0... the LX2201 chip reports whether or not it is currently charging the battery to the processor on a GPIO
21:04:20Bgerare we reading this info?
21:04:53Bgerand how to check this in gerenal...
21:04:54lostlogicit's checked every loop through the power_thread_sleep function in powermgmt.c IIRC
21:06:04lostlogicHardware or software sense? Hardware-wise will be very device specific, if we aren't controlling the charge, the charge chip would have to report it. Software-wise, I stuck a method somewhere for it... check the gwps-common.c section for bc
21:06:23***Saving seen data "./dancer.seen"
21:06:24 Quit linuxstb_ ("CGI:IRC")
21:07:11Bgerno no, i'm interested in the software part
21:07:28Bgerk, 10x
21:08:22linuxstbamiconn: Yes, the irony of alac not working on the ipod hasn't escaped me. I'm planning on trying the profiling code on it soon.
21:10:10Learsafetydan: hm... eq_gain_menu isn't that translator friendly, as the string buffers are fairly small...
21:12:19safetydantrue, but I picked that number based on the English length. I should probably double it
21:13:16Learalso, the "5" in the "%5d" formatting doesn't quite work, as a space and a digit isn't equally wide in the font I use... :)
21:13:57linuxstbjjoao: Yes, that video is Rockbox on an iPod photo.
21:14:22safetydanWhat? You mean the whole world doesn't use monospace fonts?
21:17:38 Join bluey [0] (
21:23:52jjoaolinuxstb: It is only missing the WSP screen
21:24:48linuxstbjjoao: The WPS screens can be seen on the Rockbox Wiki (WPSGallery) or the H300 forums. The H300 LCD is identical to the iPod LCD - so the same WPSs will work.
21:26:00LearBtw, regarding ipods... How is the character spacing on them? Is it like in nimbus/chicago in rockbox, with two pixels between chars?
21:27:03jjoaolinuxstb: How about the battery life indicator? is it possible configure the WSP to scrool the battery life on ipod?
21:30:02 Join linuxstb_ [0] (
21:30:14 Quit linuxstb (Nick collision from services.)
21:30:16 Nick linuxstb_ is now known as linuxstb (
21:30:21*amiconn is testing linuxstb's backdrop pointer patch, modified to work with 'const' attributes and some more slowness removed :)
21:30:46lostlogicamiconn: I think you may be more of a performance whore than me :)
21:31:05amiconnlinuxstb: Btw, lcd_clear_display() doesn't work correctly with backdrop enabled
21:31:32linuxstbIn what way?
21:32:01amiconnIf DRMODE_INVERSEVID is set, lcd_clear_display() should fill with the foreground colour
21:32:44amiconnApart from that, clearing to the backdrop is dead simple - just a memcpy()
21:36:30amiconnlinuxstb: The declaration for lcd_fastpixelfuncs looks rather weird if I want to keep the const for the arrays:
21:36:39amiconnextern lcd_fastpixelfunc_type* const * lcd_fastpixelfuncs;
21:36:59 Join elinenbe [0] (
21:37:22amiconn'a pointer to a constant array of pointers to lcd_fastpixelfunc_type functions'
21:39:42 Quit gtkspert_laptop (Read error: 110 (Connection timed out))
21:40:21 Quit Therx (Read error: 104 (Connection reset by peer))
21:41:18jjoaolinuxstb: How about the battery life indicator? is it possible configure the WSP to scrool the battery life on ipod?
21:42:43peturjjoao: afaik, battery stuff isn't ready on ipod
21:45:01jjoaois it in progress or its not important at moment?
21:45:20amiconnNow on to testing...
21:45:26amiconnBackdrop image, anyone?
21:45:43 Join Therx [0] (
21:46:31Papricame me ;]
21:46:33peturjjoao: they're quite busy, one thing at the time.. be patient
21:46:59 Join Musicmad [0] (
21:47:16 Quit Musicmad (Client Quit)
21:50:03 Join amiconn_ [0] (
21:50:18 Quit amiconn (Nick collision from services.)
21:50:19 Nick amiconn_ is now known as amiconn (
21:50:57*amiconn wonders why linuxstb puts the seldom-used lcd_backdrop in iram, but the often-used lcd_backdrop_offset in dram...
21:54:11 Join slimx [0] (
21:55:04 Join linuxstb_ [0] (
21:56:08Papricacan someone take a look on that
21:56:32Papricaand tell me if he know what is wrong with the end of the first line?
21:56:32 Quit linuxstb (Read error: 110 (Connection timed out))
21:56:48 Join DreamTactix291 [0] (
21:58:44 Join solexx [0] (
22:01:16peturbecause it isn't monospaced, you're sometimes writing past the margin?
22:01:20_FireFly_Paprica: what do you mean with worng with the last lane
22:01:33Midgey34look at the right side o
22:01:41Midgey34f the line
22:01:45*petur hands _FireFly_ his contactlenses
22:02:00Papricapetur, so what can i do?
22:02:18_FireFly_oops now i see :)
22:03:00peturyou're implementing margins for scrolling text? haven't looked at that part of the code yet
22:03:26peturhow are you limiting the output?
22:03:39amiconnPaprica: Your patch doesn't handle right-cropped letters correctly
22:04:14amiconnThe old code didn't need to, since it used the physical right margin
22:04:36jjoaopetur: ok, :) thx
22:04:55Papricai try to limit it more, but it did the same
22:05:17Papricaamiconn, so i need to write new one?
22:05:48amiconnI don't know how your code works, but cvs code does only left-cropping
22:06:14peturclear the area past the last letter?
22:06:54Papricapetur, this is what i did in the old code.
22:06:54amiconnMay be difficult. Other content could already be rendered there
22:08:28Learyou'll probably need to calculate a mask and apply that to the last column(s).
22:08:42peturme admits he has no idea what he's talking about
22:09:33peturme and this bloody keyboard ;)
22:11:30amiconnNo mask needed. Just track the consumed width and limit the font bitmap output to that if it's smaller than the letter width
22:12:46amiconnlcd_putsxyofs() tracks width anyway
22:12:47 Join linuxstb [0] (
22:13:08amiconnlinuxstb: Check cvs :)
22:13:13LearAh, yes, the bitmap is rotated here...
22:13:52linuxstbamiconn: Linus and I discussed the idea of viewports the other day - so you would do something like lcd_set_viewport(x,y,width,height) and all future text output (and graphics?) would be rendered inside that viewport.
22:14:28 Quit solexx_ (Read error: 110 (Connection timed out))
22:16:05 Join amiconn_ [0] (
22:16:20linuxstbamiconn: lcd-16bit.c looking good again. I assume that the speed for a normal bgcolor is the same as it was before the backdrop patch?
22:16:22 Quit amiconn (Nick collision from services.)
22:16:22 Nick amiconn_ is now known as amiconn (
22:16:33amiconnIt should
22:17:02amiconnIn fact, lcd_clear_display() might be even faster than without
22:17:13amiconnThat will change with memset16() of course
22:17:35amiconnlcd_clear_display() might be even faster *with backdrop* than without
22:19:38 Join Rondom [0] (
22:20:51*amiconn noticed an unwanted side effect of backdrops
22:21:27amiconnThe backdrop replaces *all* background pixels, even where this isn't wanted
22:21:54linuxstbWhat do you mean?
22:22:00amiconn-> e.g. splashes loose their grey background
22:22:34peturI loved that effect - it isn't intentional?
22:22:49linuxstbOne person's bug, is another's feature....
22:22:57*Lear likes the equalizer.
22:23:41linuxstbamiconn: Is there anything apart from the splash?
22:23:55peturbtw, I didn't see the backdrop when running a plugin, do I miss something?
22:24:03amiconnIt's the only one I discovered so far...
22:24:32linuxstbpetur: That's intentional (for now). Some plugins (like cube) would be cool with the backdrop, but not every plugin.
22:24:58*safetydan has the beginnings of a graphic (i.e. with sliders) EQ UI
22:24:59linuxstb(I changed plugin.c to revert to the bgcolor before starting the plugin)
22:25:03 Join Paco_ [0] (
22:25:06*amiconn will try to colorify some plugins
22:25:21amiconnThinking about mandelbrot, and demystify
22:25:59linuxstbI re-discovered star the other day - that could be a nice game if it was adapted for the larger LCDs.
22:26:45peturI accidently ran start today and it hung my H340 - had to reset it
22:26:47linuxstbbouce could do with a colour logo as well.
22:27:30peturtyping has never been better ;)
22:28:11Paco_hello there, I'd just like to thank and congratulate all those who contributed to rockbox for this amazing piece of software ! my H120 feels at least 10x better now ! :)
22:32:15amiconnbounce is in need for fixing the background clock
22:32:21amiconn*in need of
22:36:06*amiconn needs suggestions for mandelbrot button assignments on iPod
22:36:39Therxnano ipod resolution anyone>
22:37:11 Quit Paco_ (Remote closed the connection)
22:38:01amiconnI need left/right/up/down, zoom in/out, reset to defaults, increase/decrease # of iterations, quit
22:38:45linuxstbKeep wishing....
22:39:03Therxnoone? ok
22:39:14amiconnlinuxstb: It's possible on Ondio...
22:39:42 Part LinusN
22:40:07 Join LinusN [0] (
22:40:38BagderLinusN: that was a mighty score!
22:41:50linuxstbup/down/left/right can menu/play/left/right. Zoom in/out will be scroll fwd/back, select+play can be reset to defaults, select+left/right for #iterations, and select+menu is quit
22:42:18linuxstbAnd I think that's every combination taken....
22:42:50linuxstbTherx: Also look in firmware/export/config-ipodnano.h
22:43:59 Join tomal [0] (
22:45:47 Quit Rondom (No route to host)
22:46:53tomalHas anyone been using DEBUG builds recently? apps/Makefile seem not to build rockbox.elf with DEBUG enabled.
22:47:00tomalOr am I missing something?
22:47:44 Join Rondom [0] (
22:48:01BagderI think only LinusN ever did use them and that he hasn't used them in a while
22:49:05tomalOk, so I will fix apps/Makefile so that it builds rockbox.elf when DEBUG is enabled.
22:51:30 Join vektek [0] (
22:51:41 Quit Therx (Read error: 104 (Connection reset by peer))
22:52:48 Quit vektek (Client Quit)
22:55:30imphasingslimx: Ping?
22:56:50Bgerlostlogic ?
22:59:12 Quit darkless (Read error: 104 (Connection reset by peer))
22:59:39 Join darkless [0] (
23:01:18 Part safetydan ("Leaving")
23:06:27***Saving seen data "./dancer.seen"
23:07:59*amiconn has to implement high-color scrolling for mandelbrot
23:08:19amiconn...and I still didn't finish memmove() :-(
23:08:34 Quit Bger ("[BX] Have YOU raped a lawn gnome today?")
23:09:14peturis there anything checking the config settings read from disk against their limits?
23:09:56lostlogicBg3r: here
23:10:35peturhe just left...
23:10:48Midgey34tomal: can you take a look at my blackjack plugin and come up with some good keybindings?
23:12:13tomalMidgey34: I have to take a look at all the plugins with missing ifp keybindings...
23:12:58tomalMidgey34: I will look at yours now.
23:13:45 Quit linuxstb ("CGI:IRC")
23:13:49 Join linuxstb [0] (
23:16:15Midgey34my current source varies from whats on the website and I actually have keybindings that will compile, I just think they may be awkward
23:17:28BHSPitMonkeymy bleeding edge installation from last night is really crazy
23:17:53BHSPitMonkeyhave there been any bugs in the cvs?
23:18:34Midgey34whats happening?
23:19:19amiconnH3x0 runtime isn't that shabby. 12:40 now, still running
23:19:34amiconn(mp3 plus some bejeweled)
23:20:36Midgey34amiconn: what bitrate?
23:20:49amiconnlame −−preset standard, and some cbr 128
23:21:03Midgey34not bad
23:21:50tomalMidgey34: I suggest that you assign up/down/left/right and select (joystick press on ifp) in a similar way as on other targets.
23:22:01LearHrm, the explorer is supposed to restart when it crashes... :(
23:22:32tomalMidgey34: Use BUTTON_PLAY for quit (it is used to exit from menus in the original firmware)
23:23:12slimximphasing, echo reply :)
23:23:35tomalMidgey34: and BUTTON_MODE for save/resume (it is also a recording button on ifp)
23:24:07imphasingslimx: I was just wonder how the 3G development goes, and if you have a patch?
23:24:11imphasingSans the ATA probably.
23:24:19amiconnH3x0 just shut down...
23:24:45Midgey34tomal: if it unclear, BJACK_SELECT is the highscores
23:24:48slimxi do
23:24:59slimxbut still ata ....
23:25:02Midgey34I'll probably change the name
23:25:16imphasingI have the one from sourceforge
23:25:23imphasingand I'm looking at the iPL ata code
23:25:31slimxi can commit it in sf
23:25:35 Join Myth`n`Legend [0] (
23:25:48slimxwait a minute please i diff it
23:26:00imphasingOk, cool.
23:26:26linuxstbslimx: Yes, please update the sourceforge patch - I can have a look as well to see if I can see why the ata driver isn't working.
23:26:56imphasingThe ATA code for iPL I think is in ipod-ide.c, but I'm not sure..
23:27:12imphasingEh, maybe not.
23:27:25linuxstbimphasing: Most of the ata code is just the standard uclinux ata driver. Only the initialisation code is specific to ipodlinux.
23:28:01tomalMidgey34: You may also use BUTTON_EQ for something (it is located between play and mode on the player)
23:28:24linuxstbI'm guessing that it's something else in Rockbox causing the problem, rather than the ata driver itself - the differences between the pp5002 and pp5020 ata drivers are tiny.
23:28:44tomalSomeone (me?) will have to take a look at all the ifp keybindings anyway.
23:29:02imphasingYeah, I figured it would be fairly trivial to port to 3G, just need a greyscale driver, and change out some CPU code.
23:29:27imphasingslimx: Do you have a 3G build option in the makefile yet?
23:29:34Midgey34I'm using BUTTON_EQ for stay right now
23:30:09 Join kkurbjun [0] (
23:30:17slimxbootloader only
23:30:26imphasingAh, cool.
23:30:53 Quit Lear ("Chatzilla 0.9.70 [Firefox]")
23:31:03Midgey34kkurbjun: new doom patch seems promising
23:31:21Midgey34haven't had an opportunity to try it however
23:32:00kkurbjunmidgey34: yes, I found it pretty much playable. There's still room for improvement though
23:32:16muesli__fine job kkurbjun :)
23:32:19 Quit petur ("here today, gone tomorrow")
23:32:30kkurbjunthank you
23:33:14kkurbjunmuesli: have you tried the latest?
23:33:19 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
23:33:36muesli__when did u update latest?
23:33:38*Myth`n`Legend lurks in his corner
23:33:47kkurbjunearlier today
23:34:19muesli__in hours? its 11.30pm over there ;)
23:34:26 Quit linuxstb ("CGI:IRC")
23:34:35slimximphasing, the patch is in sf
23:34:42imphasingOk, thanks
23:35:01kkurbjunumm, probably 6 hours ago
23:35:04slimxthe bootloader main code is in bootloader/ipod.c
23:35:40slimxbuild target is ipod 3G i guess
23:35:41 Quit NicoFR ()
23:36:08 Nick linuxstb_ is now known as linuxstb (
23:36:42 Quit aegray (Read error: 104 (Connection reset by peer))
23:36:57slimxi think that some ata patterns might wrong for the ipod 3G
23:37:51imphasingslimx: Ok, I'll try it out.
23:38:31 Join aegray [0] (
23:38:33slimxlinuxstb, ok
23:38:41imphasingErm..where do I need to put that patch when I apply it?
23:38:45imphasingIt's giving me a few errors.
23:39:35slimxin the source directory
23:40:06slimxit's an eclipse made one
23:40:16slimxcheck paths in the patch
23:40:25imphasingyeah, I'm doing that
23:41:08slimxyes that's in the source root directory
23:41:23linuxstbslimx: The patch doesn't apply cleanly against current cvs.
23:41:46slimxi must do a cvs update so
23:41:48linuxstb - problems in ata.c and power.c
23:42:28imphasingYeah, I get a failed hunk too
23:42:47 Quit aegray (Read error: 104 (Connection reset by peer))
23:43:21slimxthe ata driver is the same as in cvs
23:43:43slimxso if it don't apply replace it with current
23:43:54slimxi'm updating my source tree
23:44:17 Quit bluey ("Leaving")
23:44:40 Join DrMoos [0] (
23:44:50 Quit Moos (Read error: 104 (Connection reset by peer))
23:44:53 Join bluey [0] (
23:45:05 Nick DrMoos is now known as Moos (
23:45:14linuxstbslimx: I've found one important bug - in your definition of udelay (firmware/export/system.h), you shouldn't use inl(TIMER1_VAL), you just use TIMER1_VAL
23:45:21 Join aegray [0] (
23:45:49 Quit bluey (Client Quit)
23:46:26 Join bluey [0] (
23:47:39 Quit _FireFly_ ("Leaving")
23:47:46slimxi've put a new patch
23:47:59slimxafter updating the source
23:48:35slimxi'm looking system.h
23:49:06imphasingI'm getting a load of file not found errors..
23:49:12imphasingbut it seems to be at the right place
23:50:03imphasingyou have your source set up like cvsroot/rockbox/?
23:50:41slimxi use src-all repository
23:51:01amiconnSomething is wrong with lcd_update_rect() on H3x0
23:51:17 Quit aegray (Read error: 104 (Connection reset by peer))
23:51:21amiconnIf I call that with a width of 1, the H340 freezes
23:52:42linuxstbslimx: The new files in your patch are not correct. Your first patch worked though.
23:53:10 Join aegray [0] (
23:54:06slimxthere is cvs merge error on power.c
23:54:18slimxin the patch
23:54:51kkurbjunamiconn, is it possible/ is it acceptable for plugins to access the video memory directly and handle their own buffering?
23:55:13 Join Rondom_ [0] (
23:55:13 Quit Jolt|Away ("CGI:IRC (EOF)")
23:55:34amiconnYou can't access the video memory directly. It is built into the LCD driver chip, and not memory mapped
23:56:39kkurbjunwould it be possible to copy it directly from a plugin buffer to the LCD memory though?
23:56:39 Quit aegray (Read error: 104 (Connection reset by peer))
23:56:54 Quit Rondom (No route to host)
23:57:01amiconnYes, but what's the point?
23:57:20kkurbjunthe reason I'm asking is both rockboy and doom have their own buffers effectively, so having another intermediate buffer seems a bit of overkill
23:57:24amiconnJust render into the framebuffer and call rb->Lcd_update() afterwards. Exactly the same
23:57:28lostlogicSlasheri: (or anyone) I cannot for the life of me figure out why this patch causes a packet worth of old audio to play during the new track on most manual skips... otherwise, the patch is ready to be committed as a better pcmbuf structure
23:58:37amiconnSomething is fishy...
23:58:56 Join aegray [0] (

Previous day | Next day