00:00:45 | | Quit n1s () |
00:00:47 | | Quit jhMikeS (Nick collision from services.) |
00:00:53 | | Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS) |
00:05:25 | | Quit {phoenix} ("Konversation terminated!") |
00:05:45 | *** | Saving seen data "./dancer.seen" |
00:07:23 | | Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]") |
00:08:28 | | Join miepchen^schlaf [0] (n=miepchen@p579EC9A6.dip.t-dialin.net) |
00:14:13 | | Quit bluebrother ("leaving") |
00:16:48 | | Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon) |
00:17:29 | | Quit bertrik ("Leaving") |
00:30:55 | | Quit n17ikh () |
00:33:34 | | Join PaulJam__ [0] (n=PaulJam_@vpn-3009.gwdg.de) |
00:36:30 | | Join robin0800 [0] (n=robin080@cpc2-brig8-0-0-cust394.brig.cable.ntl.com) |
00:39:27 | | Join Thundercloud [0] (n=thunderc@84-51-130-71.judith186.adsl.metronet.co.uk) |
00:39:42 | | Quit ender` (" Join the army, meet interesting people, kill them.") |
00:41:22 | | Join kushal_12_27_200 [0] (n=kushal@12.169.180.178) |
00:42:28 | | Quit rasher (Read error: 110 (Connection timed out)) |
00:46:38 | | Join PaulJam [0] (n=PaulJam_@vpn-3009.gwdg.de) |
00:51:25 | | Quit Schmogel ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") |
00:51:58 | | Quit PaulJam_ (Read error: 113 (No route to host)) |
00:54:47 | | Quit culture (Read error: 110 (Connection timed out)) |
00:58:24 | | Quit fcr (Read error: 110 (Connection timed out)) |
01:00 |
01:01:14 | | Quit PaulJam__ (Read error: 113 (No route to host)) |
01:01:17 | | Join bughunter2 [0] (n=Jelle@77.164.66.126) |
01:04:48 | | Quit PaulJam (".") |
01:08:32 | | Quit BHSPitLappy (Remote closed the connection) |
01:09:03 | kushal_12_27_200 | hey JdGordon, have you noticed the "refreshing database" problem in any sansa c250 recently? The OF in mine is locked forever in that screen and the only way I am charging it is by pressing select when connecting it. |
01:17:44 | | Join Slack [0] (n=brett@12-218-63-169.client.mchsi.com) |
01:20:29 | JdGordon | I dont have a c200 |
01:23:44 | pixelma | I have one but didn't notice any problems like this. Also on my c200 the database refreshes when I disconnect USB not when I connect |
01:28:34 | | Quit perrikwp ("http://www.mibbit.com ajax IRC Client") |
01:33:31 | | Quit Thundercloud (Read error: 104 (Connection reset by peer)) |
01:34:52 | kushal_12_27_200 | Pixelma,that is what normally happens with mine too. However, it has been acting a bit strangely lately. I had to press power button for over ten seconds for it to turn off because the refresh was taking too long. It was over an hour of waiting and I finally went ahead and turned it off by holding the power button. Now, it wants to refresh database when I turn it on. |
01:36:52 | kushal_12_27_200 | JdGordon, I noticed you closed something on filespray for a (seemingly) related incident. (Do you want me to find it?) |
01:42:16 | pixelma | from your description I would suspect a problem with the OF (maybe filesystem ones too), having read somewhere that the OF's database refresh can choke on some files too. My best guess though and I'm off now |
01:43:48 | | Part pixelma |
01:44:04 | | Quit robin0800 (Read error: 104 (Connection reset by peer)) |
01:44:34 | | Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma) |
01:49:10 | | Quit HellDragon (Client Quit) |
01:50:31 | | Join fcr [0] (n=fran@2001:470:89a5:0:214:a5ff:fea5:4bef) |
01:55:12 | | Join nuonguy [0] (n=john@c-71-198-1-139.hsd1.ca.comcast.net) |
01:55:52 | | Join HellDragon [0] (n=jd@modemcable100.136-203-24.mc.videotron.ca) |
01:57:06 | | Quit fcr (Nick collision from services.) |
01:57:07 | | Join fcr_ [0] (n=fran@2001:470:89a5:0:214:a5ff:fea5:4bef) |
02:00 |
02:01:20 | | Quit DerDome ("Leaving.") |
02:01:49 | | Join Strife89 [0] (n=michael@204.116.245.152) |
02:05:03 | | Join bughunter21 [0] (n=Jelle@77.164.66.126) |
02:05:47 | *** | Saving seen data "./dancer.seen" |
02:07:44 | | Join rasher [0] (n=rasher@0x5550f5a3.adsl.cybercity.dk) |
02:08:36 | | Join hd [0] (n=jd@modemcable100.136-203-24.mc.videotron.ca) |
02:08:57 | | Quit HellDragon (Read error: 104 (Connection reset by peer)) |
02:13:27 | | Quit bughunter2 (Read error: 110 (Connection timed out)) |
02:19:12 | | Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/mibbit.com/x-7104536124660e35) |
02:25:46 | | Quit Strife89 ("Bye, guys!") |
02:26:50 | | Join AhtiK [0] (n=ahti@ip67-152-80-226.z80-152-67.customer.algx.net) |
02:29:12 | | Quit MethoS-- (Remote closed the connection) |
02:29:26 | | Nick hd is now known as HellDragon (n=jd@Wikipedia/HellDragon) |
02:37:03 | | Quit rasher (Read error: 104 (Connection reset by peer)) |
02:37:41 | | Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey) |
02:40:13 | | Nick fcr_ is now known as fcr (n=fran@2001:470:89a5:0:214:a5ff:fea5:4bef) |
02:43:07 | | Join Darksair [0] (n=user@221.221.165.108) |
02:44:45 | | Quit Slack ("Ex-Chat") |
02:49:46 | | Quit kugel ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]") |
02:52:29 | | Join reacocard_ [0] (n=reacocar@WL-112.CINE.HMC.Edu) |
02:53:05 | | Quit Nibbl (Read error: 113 (No route to host)) |
02:53:09 | | Quit reacocard (Nick collision from services.) |
02:53:14 | | Nick reacocard_ is now known as reacocard (n=reacocar@WL-112.CINE.HMC.Edu) |
03:00 |
03:02:25 | | Join JdGordon|uni [0] (i=768aa359@gateway/web/ajax/mibbit.com/x-c6adf282a9a7276b) |
03:03:09 | JdGordon|uni | kushal_12_27_200: which task did I close related to the c200 of db? |
03:07:25 | kushal_12_27_200 | http://www.rockbox.org/tracker/task/7207 , not sure how closely related ... |
03:08:53 | JdGordon|uni | wow, old report... |
03:09:20 | kushal_12_27_200 | more than a year old |
03:10:18 | kushal_12_27_200 | do you think it is significant at all? |
03:11:15 | JdGordon|uni | well that report was for the e200, not c200... on the e the DB refresh happens before usb connect |
03:11:27 | JdGordon|uni | apparently the c does it after usb disconnect, so there is nothing we can do about it |
03:16:10 | | Quit CaptainKewl ("( www.nnscript.de :: NoNameScript 4.02 :: www.XLhost.de )") |
03:16:51 | kushal_12_27_200 | JdGordon, is it possible to entirely bypass the OF |
03:17:06 | JdGordon|uni | not untill we get usb working |
03:18:20 | kushal_12_27_200 | hmm |
03:19:40 | kushal_12_27_200 | is there a bug report filed for that yet? |
03:19:53 | JdGordon|uni | for what? |
03:20:13 | kushal_12_27_200 | libusb |
03:20:23 | | Quit num1 ("veni, vidi et capiebar ad anum "I came, I saw and I endured a rear assault"") |
03:20:42 | JdGordon|uni | libusb is a linux thing.... the lack of usb in rockbox is well known... |
03:23:07 | kushal_12_27_200 | how could we fix that? |
03:24:11 | | Quit HellDragon (Remote closed the connection) |
03:25:24 | | Join HellDragon [0] (n=jd@modemcable100.136-203-24.mc.videotron.ca) |
03:25:57 | | Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-64ae5c09e6e94201) |
03:26:29 | | Join CaptainKewl [0] (n=jason@207-237-173-165.c3-0.nyr-ubr4.nyr.ny.cable.rcn.com) |
03:26:49 | kushal_12_27_200 | JDGordon, can I delete files in the player using rockbox itself? |
03:26:58 | JdGordon|uni | yes |
03:27:06 | kushal_12_27_200 | how would I do that? |
03:28:01 | kushal_12_27_200 | I want to delete everything inside the folder /Music |
03:28:14 | JdGordon|uni | read the manual |
03:28:37 | kushal_12_27_200 | sure thing |
03:30:25 | kushal_12_27_200 | long select does it, thanks |
03:30:55 | | Quit jhulst (Remote closed the connection) |
03:32:39 | | Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net) |
03:37:14 | | Quit kushal_12_27_200 ("Leaving") |
03:41:14 | | Join EspeonEefi [0] (i=espeonee@STRATTON-SEVEN-TWENTY-ONE.MIT.EDU) |
03:44:45 | | Join Slack [0] (n=brett@12-218-63-169.client.mchsi.com) |
03:50:50 | | Quit obo ("bye") |
04:00 |
04:03:28 | | Join mab1376 [0] (n=mab1376@pool-96-246-29-241.nycmny.fios.verizon.net) |
04:03:34 | mab1376 | hello |
04:04:10 | mab1376 | im trying to enable the database on my gigabeat and i cannot figure out what the yes buttoin is to enable it for the life of me |
04:04:29 | | Join jhulst [0] (n=jhulst@unaffiliated/jhulst) |
04:05:34 | JdGordon|uni | which gigabeat? |
04:05:46 | JdGordon|uni | either way, it should be the usual "select" button |
04:05:47 | mab1376 | F40 |
04:05:48 | *** | Saving seen data "./dancer.seen" |
04:06:24 | JdGordon|uni | you shold ckeck the manual... |
04:06:30 | mab1376 | bare in mind i just go this in the mail today, so i have no idea which button is which. |
04:06:49 | mab1376 | good idea |
04:06:49 | JdGordon|uni | and yes, its the select button |
04:06:54 | JdGordon|uni | whichever that is |
04:07:21 | mab1376 | hmm ok thanks |
04:07:34 | JdGordon|uni | but go read the manual anyway |
04:09:53 | mab1376 | ok |
04:15:22 | | Quit EspeonEefi ("ăăăȘă") |
04:15:55 | | Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/mibbit.com/x-d46ffb8abd1abfb9) |
04:20:03 | | Quit CaptainKewl (Read error: 110 (Connection timed out)) |
04:20:06 | | Join blkhawk- [0] (n=blkhawk@e179050024.adsl.alicedsl.de) |
04:29:18 | | Quit mab1376 ("I-n-v-i-s-i-o-n 3.0 (March '08)") |
04:30:35 | | Quit fdinel ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") |
04:36:27 | | Quit blkhawk (Read error: 113 (No route to host)) |
04:36:59 | | Join miepchen^schlaf_ [0] (n=miepchen@p579ECEF9.dip.t-dialin.net) |
04:37:04 | | Nick blkhawk- is now known as blkhawk (n=blkhawk@e179050024.adsl.alicedsl.de) |
04:49:21 | | Join EspeonEefi [0] (i=espeonee@SYDNEYPACIFIC-THREE-ELEVEN.MIT.EDU) |
04:51:54 | | Quit miepchen^schlaf (Read error: 110 (Connection timed out)) |
05:00 |
05:01:08 | | Quit saratoga ("CGI:IRC (EOF)") |
05:03:57 | | Quit JdGordon|uni ("http://www.mibbit.com ajax IRC Client") |
05:45:22 | | Quit intrados (Read error: 110 (Connection timed out)) |
05:48:12 | | Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.") |
05:48:18 | | Quit nplus (Read error: 104 (Connection reset by peer)) |
06:00 |
06:04:27 | | Join homielowe [0] (n=homielow@d206-116-134-81.bchsia.telus.net) |
06:05:49 | *** | Saving seen data "./dancer.seen" |
06:11:51 | | Join kushal_12_27_200 [0] (n=kushal@12.169.180.178) |
06:15:35 | | Quit jhulst (Remote closed the connection) |
06:30:43 | | Quit Darksair ("Use the Force, Luke!") |
06:31:00 | | Join AndyIL [0] (i=AndyI@212.14.205.32) |
06:44:48 | | Quit AndyI (Read error: 110 (Connection timed out)) |
06:48:42 | | Quit Zarggg () |
07:00 |
07:03:16 | | Quit reacocard (".") |
07:03:40 | | Join bmbl [0] (n=Miranda@unaffiliated/bmbl) |
07:03:43 | | Join reacocard [0] (n=reacocar@134.173.59.155) |
07:13:42 | | Join Nibbl [0] (n=Nibbler@91-66-158-199-dynip.superkabel.de) |
07:18:50 | | Join J-23 [0] (n=kvirc@a105.net128.okay.pl) |
07:27:22 | | Quit JdGordon ("Konversation terminated!") |
07:30:12 | | Join syn4pse [0] (n=syn4pse@cpe-075-176-144-005.sc.res.rr.com) |
07:31:52 | | Join JdGordon [0] (n=Miranda@c211-28-145-137.smelb2.vic.optusnet.com.au) |
07:32:51 | syn4pse | i'm running into trouble using 2 arrays of 6400 char, even though it is defined globally. this will be a full array (a maze). is this due to a limitation in memory? |
07:33:14 | syn4pse | *they are , that is |
07:33:28 | JdGordon | what sort of trouble? |
07:33:48 | syn4pse | well, when i access the full array, i crash pretty miserably |
07:34:02 | syn4pse | but if i just use a portion of the array(s), i am fine |
07:34:20 | JdGordon | thats not a very big array, so assuming your accessing it correctly there shouldnt be a probleem |
07:34:29 | syn4pse | i must be handling it wrong |
07:34:53 | syn4pse | i'll keep on banging away at it. |
07:35:26 | syn4pse | another thing, it only crashes on my ipod, not on the simulator |
07:36:39 | syn4pse | so i'm not close to a limit? that's only 6400*1*2, 12800 bytes, right? |
07:36:56 | | Quit fcr (Read error: 104 (Connection reset by peer)) |
07:37:04 | cool_walking_ | Llorean told me the plugin buffer is 512K on iPod Video |
07:37:28 | | Quit J-23 ("Flying cow pressed ^D on my keyboard.") |
07:37:28 | syn4pse | seems like i'm well within that, so i must be missing something else. |
07:37:52 | Llorean | What sort of crash? |
07:38:10 | syn4pse | the screen flashes and the ipod powers down |
07:39:52 | Llorean | And you've gone through step by step to figure out which line, exactly actually crashes on? |
07:40:14 | syn4pse | i can't crash it on the simulator, unfortunately |
07:40:17 | syn4pse | not for lack of trying |
07:40:33 | syn4pse | i'd love to crash it in gdb :) |
07:40:44 | Llorean | You don't need GDB to go line by line. |
07:40:55 | syn4pse | i'm a little new at this |
07:41:17 | Llorean | You could guess where you think it is, put a prompt for a keypress in several places that also display a recognizable string, see how many strings you get to see before the crash |
07:41:22 | Llorean | Then narrow it down similarly |
07:41:33 | Llorean | It's not ideal, and if you've got a lot of iterations it could take a very long time. |
07:41:52 | Llorean | But if you honestly can't figure it out by analyzing the current logic it at least gives you a way to prod at it. |
07:42:00 | syn4pse | i have an idea where i think it is but i can't seem to figure out my logical error. |
07:42:05 | syn4pse | it's frustrating :) |
07:42:41 | syn4pse | i'll try that |
07:42:46 | | Quit kushal_12_27_200 ("Leaving") |
07:45:43 | syn4pse | it has to be bounds checking on the array. i think i have a hole in there somewhere. |
07:50:34 | | Quit Nibbl (Read error: 60 (Operation timed out)) |
07:50:48 | | Join Nibbl [0] (n=Nibbler@91-66-158-199-dynip.superkabel.de) |
07:57:11 | | Join Darksair [0] (n=user@221.221.165.108) |
08:00 |
08:02:23 | | Quit syn4pse ("Leaving") |
08:02:57 | | Join flx [0] (i=flux@jolt.modeemi.cs.tut.fi) |
08:05:50 | *** | Saving seen data "./dancer.seen" |
08:06:46 | | Join Jabone_ [0] (i=jpylvana@157.24.184.130) |
08:07:23 | | Quit flux (Read error: 104 (Connection reset by peer)) |
08:07:36 | | Quit miepchen^schlaf_ () |
08:07:54 | | Nick flx is now known as flux (i=flux@jolt.modeemi.cs.tut.fi) |
08:08:04 | | Quit Jabone (Read error: 104 (Connection reset by peer)) |
08:09:47 | | Quit bmbl (Read error: 104 (Connection reset by peer)) |
08:12:33 | | Quit JdGordon (Read error: 104 (Connection reset by peer)) |
08:14:36 | | Nick bughunter21 is now known as bughunter2 (n=Jelle@77.164.66.126) |
08:15:14 | | Join JdGordon [0] (n=jonno@c211-28-145-137.smelb2.vic.optusnet.com.au) |
08:16:02 | | Quit perrikwp ("http://www.mibbit.com ajax IRC Client") |
08:17:26 | | Quit gevaerts (Nick collision from services.) |
08:17:38 | | Join gevaerts [0] (n=fg@rockbox/developer/gevaerts) |
08:26:02 | | Quit Darksair ("Everything that has a beginning has an end.") |
08:26:23 | | Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/mibbit.com/x-9b3502e601643224) |
08:26:35 | | Join Darksair [0] (n=user@221.221.165.108) |
08:29:12 | | Join ender` [0] (i=krneki@foo.eternallybored.org) |
08:31:48 | | Join syn4pse [0] (n=chatzill@cpe-075-176-144-005.sc.res.rr.com) |
08:34:25 | | Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl) |
08:37:07 | | Join GodEater [0] (i=c2cbc962@gateway/web/ajax/mibbit.com/x-17ec5247e509dc03) |
08:37:54 | | Quit Slack ("Ex-Chat") |
08:39:58 | | Quit bughunter2 ("bye") |
08:41:53 | | Quit BigBambi (Read error: 113 (No route to host)) |
08:43:58 | | Quit reacocard (".") |
08:44:25 | | Quit Seed ("cu, Andre") |
08:45:53 | | Quit syn4pse ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]") |
08:49:17 | | Join Rob2222 [0] (n=Miranda@p4FDCE7D9.dip.t-dialin.net) |
08:56:16 | | Quit cool_walking_ ("http://www.mibbit.com ajax IRC Client") |
09:00 |
09:06:00 | | Quit GodEater ("http://www.mibbit.com ajax IRC Client") |
09:06:05 | | Join GodEater [0] (i=c2cbc962@gateway/web/ajax/mibbit.com/x-aaa91beba9e752da) |
09:07:38 | | Quit Rob2223 (Read error: 110 (Connection timed out)) |
09:12:06 | | Join BigBambi [0] (i=86ceaf40@rockbox/staff/BigBambi) |
09:21:49 | | Quit pixelma2 ("-") |
09:22:01 | | Join pixelma [50] (i=pixelma@rockbox/staff/pixelma) |
09:27:14 | | Join MethoS [0] (n=clemens@host-091-096-214-074.ewe-ip-backbone.de) |
09:31:53 | | Quit MrDuck (Read error: 113 (No route to host)) |
09:35:18 | | Quit Nibbl ("Ex-Chat") |
09:38:40 | | Quit BHSPitMonkey ("Ex-Chat") |
09:43:19 | | Join rasher [0] (n=rasher@0x5550f5a3.adsl.cybercity.dk) |
09:44:54 | | Nick Jabone_ is now known as Jabone (i=jpylvana@157.24.184.130) |
09:45:08 | | Join {phoenix} [0] (n=dirk@p54B47698.dip.t-dialin.net) |
09:51:54 | | Quit MethoS (Remote closed the connection) |
10:00 |
10:05:52 | *** | Saving seen data "./dancer.seen" |
10:08:37 | | Quit nuonguy ("This computer has gone to sleep") |
10:12:58 | | Quit AndyIL () |
10:17:33 | | Join AndyI [0] (i=AndyI@212.14.205.32) |
10:40:46 | | Join MrDuck [0] (n=kachna@r3g248.net.upc.cz) |
10:41:47 | | Quit spiorf (Remote closed the connection) |
10:46:29 | | Join petur [50] (n=petur@rockbox/developer/petur) |
10:51:24 | JdGordon | how would I go about playing the audiodebug.raw the sim can output? amarok doesnt like it |
10:52:09 | linuxstb | In Linux, the "play" command (part of sox) can do it. Or load it into any sound editor. |
10:53:59 | JdGordon | play soxio: Failed reading `audiodebug.raw': unknown file type `auto' |
10:55:52 | | Quit jhMikeS (Nick collision from services.) |
10:55:58 | | Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS) |
10:57:02 | | Join Thundercloud [0] (n=thunderc@84-51-130-71.judith186.adsl.metronet.co.uk) |
10:58:57 | JdGordon | audactiy crashes trying to open it! |
10:59:13 | * | gevaerts tells JdGordon about manpages and options :) |
10:59:15 | | Quit XavierGr () |
11:00 |
11:02:54 | linuxstb | Hint: "-t raw" |
11:12:45 | | Join pondlife [50] (n=Steve@rockbox/developer/pondlife) |
11:14:02 | | Join stoffel_ [0] (n=sfr@p57B4E8C6.dip.t-dialin.net) |
11:15:00 | | Quit MrDuck (Read error: 110 (Connection timed out)) |
11:15:53 | | Quit {phoenix} (Remote closed the connection) |
11:16:00 | | Join n1s [0] (n=nils@rockbox/developer/n1s) |
11:19:31 | | Join html-man [0] (n=d9c996ed@gateway/web/cgi-irc/labb.contactor.se/x-7949c5fe67adef24) |
11:21:13 | | Join {phoenix} [0] (n=dirk@p54B47698.dip.t-dialin.net) |
11:21:28 | | Join sunnyday[m] [0] (n=io@217.201.150.237) |
11:21:43 | | Quit html-man (Client Quit) |
11:22:02 | sunnyday[m] | hi all |
11:22:46 | sunnyday[m] | I have installed Rockbox on my Iriver h10, but ums wouldn't work automatically. Why this? |
11:23:39 | n1s | sunnyday[m]: because rockbox doesn't have usb support on that player yet |
11:24:45 | sunnyday[m] | I see, then why in the manual they say after installing Rockbox it will run in ums mode? |
11:27:07 | n1s | s |
11:27:11 | ghen | sunnyday[m]: afaik ums functionality is in the iriver firmware |
11:27:17 | ghen | I converted mine to ums before installing rockbox |
11:27:20 | ghen | (one time operation) |
11:27:58 | sunnyday[m] | how did you do that? |
11:28:12 | sunnyday[m] | (it is the 20 gb model) |
11:28:14 | n1s | sunnyday[m]: since you don't say which h10 you have i don't know exactly but the manual for thr 20 GB version clearly states that you need to press a couple of buttons to get UMS |
11:28:59 | sunnyday[m] | yes, but I thought after installing the Rockbox's bootloader I wouldn't need to press the select key anymore |
11:29:00 | | Nick Darksair is now known as Darksair-{away} (n=user@221.221.165.108) |
11:29:14 | pixelma | sunnyday[m]: can you point us exactly to the point in the manual you are referring to? |
11:29:26 | ghen | I have a 6gb model |
11:29:31 | | Quit {phoenix} (Remote closed the connection) |
11:29:39 | ghen | there's a press-this-and-that-key method for one-time UMS boot |
11:29:51 | n1s | sunnyday[m]: it says that _step 1_ is no longer needed, that is the pinhole reset |
11:30:09 | ghen | but you can permanently replace the iriver bootloader/firmware with a UMS one |
11:32:40 | | Join funman [0] (n=fun@AAnnecy-257-1-23-63.w90-28.abo.wanadoo.fr) |
11:35:25 | | Join spiorf [0] (n=spiorf@host99-208-dynamic.44-79-r.retail.telecomitalia.it) |
11:35:46 | | Quit sunnyday[m] (Read error: 104 (Connection reset by peer)) |
11:41:13 | | Join lasser [0] (n=chatzill@W97c4.w.pppool.de) |
11:50:22 | | Join {phoenix} [0] (n=dirk@p54B47698.dip.t-dialin.net) |
12:00 |
12:00:48 | | Nick Darksair-{away} is now known as Darksair (n=user@221.221.165.108) |
12:05:56 | *** | Saving seen data "./dancer.seen" |
12:15:02 | funman | Bagder: ping |
12:26:16 | | Quit {phoenix} (Remote closed the connection) |
12:34:46 | | Join J-23 [0] (n=kvirc@a105.net128.okay.pl) |
12:35:24 | | Join moos [0] (i=moos@81-66-141-133.rev.numericable.fr) |
12:35:32 | J-23 | hi! |
12:41:43 | | Quit mc2739 () |
12:46:32 | | Join intrados [0] (n=intrados@rdrt-164-107-204-170.resnet.ohio-state.edu) |
12:53:24 | jhMikeS | amiconn: pong (to yesterday's ping) |
12:56:07 | | Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net) |
12:57:40 | | Join LambdaCalculus37 [0] (n=LambdaCa@c-24-0-218-198.hsd1.nj.comcast.net) |
12:58:09 | | Quit jhMikeS (Nick collision from services.) |
12:58:15 | | Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS) |
13:00 |
13:00:05 | | Join Acky [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) |
13:16:56 | | Quit Acksaw (Connection timed out) |
13:19:42 | | Join Acksaw [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) |
13:23:11 | | Join CaptainKewl [0] (n=jason@207-237-173-165.c3-0.nyr-ubr4.nyr.ny.cable.rcn.com) |
13:23:29 | | Quit Acky (Success) |
13:23:39 | | Quit J-23 ("Flying cow pressed ^D on my keyboard.") |
13:26:28 | | Quit amiconn (Nick collision from services.) |
13:26:35 | | Join amiconn [50] (n=jens@rockbox/developer/amiconn) |
13:39:14 | | Join Seed [0] (n=ben@bzq-84-108-232-45.cablep.bezeqint.net) |
13:41:14 | | Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP) |
13:44:36 | | Join webguest58 [0] (n=d9a9e702@gateway/web/cgi-irc/labb.contactor.se/x-39bfb0c880bf82b5) |
13:47:35 | webguest58 | Question i have an sansa e280 v1 and i see the radio icon and i see the channels ( 102.3 mhz) i can search the fm band but i do'nt hear anithing ?? who can help me? |
13:50:37 | linuxstb | webguest58: Has the radio ever worked in Rockbox? Does it work in the original Sandisk firmware? |
13:52:13 | webguest58 | I had to chance the firmware to europ with fm but then it appeard on the display |
13:53:28 | linuxstb | You didn't answer either of my questions... |
13:57:09 | webguest58 | OK no with the original firmware it did'nt work so i chanched the sansa firmware to europ with fm ( no sound) then i put rockbox on my player and still no sound |
13:57:55 | Tuplanolla | you mean it didn't work even with the european firmware? |
13:59:20 | webguest58 | Yes i can see the frequentie but i can't hear it |
13:59:21 | linuxstb | webguest58: Some E200s don't have the radio chip - and will do what you describe. |
13:59:37 | linuxstb | webguest58: See here - http://www.rockbox.org/twiki/bin/view/Main/SandiskE200HardwareComponents#FM_Chip |
14:00 |
14:00:09 | | Quit Darksair (Read error: 104 (Connection reset by peer)) |
14:00:18 | | Join Darksair [0] (n=user@221.221.165.108) |
14:00:29 | webguest58 | so its possible you can see the radio icon and the frequntie's but there's no radio chip ?? |
14:01:38 | linuxstb | If you wanted to, you could open up your Sansa and have a look (or take a photo/scan of it and post it somewhere). |
14:02:09 | | Quit spiorf (Remote closed the connection) |
14:02:30 | webguest58 | oke thank you |
14:02:49 | | Join nplus [0] (n=nplus@141.25.Globcom.Net) |
14:06:00 | *** | Saving seen data "./dancer.seen" |
14:06:08 | | Quit Darksair (Read error: 54 (Connection reset by peer)) |
14:07:53 | | Quit webguest58 ("CGI:IRC") |
14:10:02 | | Join Darksair [0] (n=user@221.221.165.108) |
14:16:22 | | Join Schmogel [0] (n=Miranda@p3EE21E75.dip0.t-ipconnect.de) |
14:33:04 | | Join J-23 [0] (n=kvirc@a105.net128.okay.pl) |
14:35:31 | | Join evilnick [0] (i=0c140464@gateway/web/ajax/mibbit.com/x-1cf63a4fa4da5a51) |
14:39:21 | funman | $400 for a USB jtag adaptor made by Keil :( |
14:40:37 | funman | I should better go for the cheap parallel adaptor saratoga bought and find a USB/parallel adaptor |
14:41:06 | JdGordon | no parallel port? |
14:41:22 | funman | not on this 2 years old laptop |
14:41:35 | * | JdGordon was about to suggest a $15 pci card... |
14:42:29 | funman | there is the $129 pcmcia card |
14:42:35 | JdGordon | ouch |
14:43:37 | funman | or the ÂŁ'#REF!' (about 0$) pcmcia card |
14:43:55 | funman | ÂŁ29 on ebay |
14:45:18 | JdGordon | are there any cheapish usb kits? I've only found the really cheap parallel port ones |
14:47:17 | funman | I couldn't find any during my searches |
14:50:19 | funman | usb / parallel cable at âŹ8.34 (they are advertised only for printers however) |
14:50:38 | | Quit kushal_12_26_200 ("http://www.mibbit.com ajax IRC Client") |
14:51:15 | | Quit J-23 (Remote closed the connection) |
14:51:25 | funman | http://www.accstation.com/pdb25p1pbus1.html?acprcnetadtf081023=pdb25p1pbus1 $32.39 for a pcmcia parallel port |
14:51:34 | | Join IudeX [0] (n=52a0f8f7@gateway/web/cgi-irc/labb.contactor.se/x-7d6292cfee54fa32) |
14:52:26 | IudeX | Clip has a recovery mode? (u understand?) |
14:52:45 | funman | IudeX: no, it has no recovery mode. Only the e200v2 has one |
14:53:56 | IudeX | Ach... So I can brick my Clip? |
14:54:48 | IudeX | funman: So Can I brick this? <- now better :) |
14:56:30 | | Quit CaptainKewl (Read error: 110 (Connection timed out)) |
14:56:39 | funman | yes you can do this if you modify the mkamsboot program - but if you leave it as is you shoudn't be able to brick it |
14:57:08 | funman | because once it is bricked, you can not recover it : it's broken forever :( |
14:57:24 | funman | IudeX: where are you from ? |
14:57:46 | IudeX | funman: Poland |
14:58:07 | funman | ok, i'm french |
14:58:13 | IudeX | funman: So, I collect money on new :D |
14:58:32 | funman | you bricked your Clip ? |
14:58:40 | IudeX | fuman:: not yet |
14:59:04 | funman | don't be afraid: now that mkamsboot is fine it's hard to brick a Clip |
14:59:54 | IudeX | funman: So how can i help u on rockboxing? |
15:00 |
15:00:21 | IudeX | funman: i think, u understand me :D |
15:00:24 | funman | figuring what is missing to use the SD card |
15:00:42 | IudeX | sd card on clip? |
15:00:45 | funman | by reverse engineering the original firmware, and maybe with the help of a JTAG connection |
15:01:02 | funman | IudeX: the internal NAND flash is accessed using a SD controller; it's not a SD card you can remove |
15:01:50 | | Quit Seed ("cu, Andre") |
15:02:05 | funman | so, from a programmer point of view, it's a SD card |
15:02:27 | IudeX | yes, i'm understand |
15:04:21 | IudeX | funman: Are you working in informatics? |
15:04:33 | funman | no I'm jobless |
15:04:58 | IudeX | So you have lots of time to rockboxing :D |
15:05:03 | funman | exact ;) |
15:05:16 | linuxstb | But no food... ;) |
15:05:30 | IudeX | :) |
15:06:00 | IudeX | Maybe : Rockbox cost - 2 Euros |
15:06:21 | IudeX | It's good idea |
15:06:24 | IudeX | :) |
15:06:43 | funman | I'll try selling my GPL licensed modifications to rockbox |
15:06:59 | IudeX | nice;] |
15:07:17 | Nico_P | JdGordon: are you progressing with the PCM mixer? |
15:08:01 | IudeX | What about games on Clip - possibly no games? :) |
15:08:31 | funman | IudeX: now it's too early for games: we miss bootloader (SD/flash) and sound |
15:08:31 | JdGordon | Nico_P: no, im stuck... i can get about 30s or so of music but it sounds really bad |
15:08:38 | JdGordon | I'm not really sure what im doing :p |
15:09:36 | IudeX | funman: Sound, bl most important |
15:10:48 | Nico_P | JdGordon: could you post the patchN |
15:10:54 | Nico_P | s/N/? |
15:11:07 | JdGordon | only if you promise not to laugh :D |
15:11:10 | funman | s,s/N/?,s/N/?/, |
15:11:15 | Nico_P | ok, I promise ;) |
15:11:37 | IudeX | Are there any differences in clip and e200 series in sound? (LoL its really hard to understand me!) |
15:12:01 | funman | IudeX: I think they are using the same hardware (on the as3525 SoC) |
15:12:08 | JdGordon | Nico_P: jdgordon.info:8080/~jonno/pcmbuf.4.diff">http://jdgordon.info:8080/~jonno/pcmbuf.4.diff |
15:12:23 | Nico_P | I get 404 |
15:12:26 | IudeX | I'm know so im asking :) |
15:12:27 | JdGordon | jdgordon.info:8080/~jonno/pcmmixer.4.diff">http://jdgordon.info:8080/~jonno/pcmmixer.4.diff |
15:12:39 | Nico_P | better :) |
15:14:25 | JdGordon | there is something fundamently wrong which I am missing... there is an extra 2k in the pcm_block struct which if thats not there it corrupts the stack, I'm sure ones thats gone it will sound more correct |
15:15:01 | | Quit stoffel_ (Read error: 113 (No route to host)) |
15:15:25 | IudeX | If e200 sounds how clip, I'll buy this |
15:16:01 | JdGordon | Nico_P: arg, that patch wont compile... comment out line 1232 in pcmbuf.c after patching |
15:16:02 | | Join {phoenix} [0] (n=dirk@p54B47698.dip.t-dialin.net) |
15:16:20 | Nico_P | ok |
15:17:35 | JdGordon | jhMikeS: preglow (?): if you find yourself with nothing better to do... care to have a look at that patch and see what the probably obvious mistake is? |
15:18:50 | funman | IudeX: why do you want e200, afraid of bricking the Clip ? |
15:21:33 | linuxstb | funman: Are you now confident that both the Clip and m200v2 have 2MB+320KB RAM? And the e200v2 has 8MB+320KB? |
15:21:37 | | Join stoffel_ [0] (n=sfr@p57B4E8C6.dip.t-dialin.net) |
15:22:13 | IudeX | funman: No. Clip sounds great :P |
15:22:27 | IudeX | I'm afraid.. Some.. |
15:23:34 | funman | linuxstb: yes - but the access code was not tested on e200 |
15:24:03 | linuxstb | Will it be possible to map the two blocks of RAM together? Or will we have to split the usage? |
15:24:15 | funman | I mean not after I corrected it - but physical disassembly shows a 8Mb SDRAM chip, and the OF has a '8MB' string |
15:24:23 | funman | split |
15:24:54 | JdGordon | is 320 enough for the codec ram? |
15:25:19 | funman | the different aliases are not consecutive |
15:25:21 | linuxstb | Not currently - the codec buffer is 1MB |
15:26:34 | JdGordon | on the e200 at least thats probably not a big deal if we dont use that 320 |
15:27:03 | JdGordon | I guess the m200 could use that for plugins, low mem targets get less plugin ram anyway |
15:27:16 | funman | I bet the 320kB RAM is faster than the SDRAM |
15:27:22 | * | preglow wonders why we would possibly want to put settings in a plugin |
15:27:45 | linuxstb | funman: I was thinking that as well, and if that's the case we could treat it like IRAM on the other targets. |
15:27:52 | JdGordon | preglow: not in here please... keep that on the ml |
15:28:28 | | Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-a091448ea85e12fe) |
15:28:40 | saratoga | just reading the logs and noticed the IRAM question |
15:28:54 | JdGordon | besides.. its an idea, and one which isnt likely to happen anyway, except if we are forced to dump some bin weight |
15:28:56 | saratoga | assuming the IRAM is SRAM internally, it should be treated like IRAM on the other targets |
15:29:58 | preglow | i can never be bothered to sit down and write an ml reply |
15:30:01 | preglow | weakness of mine |
15:30:04 | preglow | never was a fan of mls |
15:30:35 | JdGordon | irc has nice instant gratification :) |
15:30:48 | saratoga | i'd like to do a 272/48 KB split between codecs and memory on the AMS chips |
15:30:54 | preglow | i just prefer instant communication |
15:30:56 | * | linuxstb tends to start replies, but then never get round to sending them, much like Rockbox ports... |
15:31:00 | preglow | i find people to be much more reasonable then |
15:31:04 | saratoga | then we could most codecs all or almost entirely in IRAM |
15:31:16 | preglow | people have a nasty tendency to be assholes over mls |
15:31:31 | preglow | not implying that you are, of course :> |
15:31:51 | JdGordon | probably because they have more time to write the reply... |
15:31:54 | preglow | saratoga: sweet lord, they have that much iram? |
15:32:06 | JdGordon | preglow: you know the pcm buf stuff dont you? |
15:32:13 | preglow | JdGordon: not very well, depends |
15:32:41 | JdGordon | if you have a minute can you have a quick look at that patch i left a few min ago and see if you can see whats wrong with it? |
15:33:02 | preglow | will see |
15:33:04 | saratoga | preglow: moores law |
15:33:12 | JdGordon | I'm trying to just take the pcm data from the dsp and dump it on the DAC with no changes, but its no worky very well |
15:33:21 | preglow | JdGordon: pcmmixer? |
15:33:27 | JdGordon | yeah |
15:33:36 | preglow | saratoga: yeah, just didn't expect it to apply to iram :) but sure, why not |
15:33:40 | funman | about the AS3525 320kB ram: 'a special kind of DRAM' 'SRAM-style interface operation' |
15:34:10 | | Quit IudeX ("CGI:IRC") |
15:34:10 | JdGordon | how much iram is there usually? |
15:34:19 | preglow | 64k-128k |
15:34:25 | funman | 20 - 65MHz operation |
15:34:34 | funman | there is 5 64KB blocks here |
15:34:45 | preglow | funman: so it's not necessarily single cycle, then |
15:34:55 | preglow | that explains the amount a bit better than moore's law |
15:35:11 | preglow | dram is way more compact than sram |
15:35:21 | funman | and way more slower ? |
15:35:25 | preglow | not necessarily |
15:35:48 | linuxstb | funman: The extra RAM also solves another problem - where in RAM the bootloader loads the main Rockbox image to. We can now simply keep the bootloader in the 320KB RAM, and load the main Rockbox image to the start of the 2MB RAM. |
15:35:48 | preglow | but most of the single cycle iram you will see is sram |
15:36:26 | funman | linuxstb: doesn't some rockbox code need to be loaded in IRAM for speed ? |
15:36:28 | saratoga | theres also 1T DRAM which can be used in place of SRAM in some situations |
15:36:45 | linuxstb | funman: Yes, but that's done by the main Rockbox crt0.S - it relocates parts of itself to IRAM. |
15:36:46 | funman | what's the '1T' for? |
15:37:14 | funman | for sure it's not 1 TeraByte |
15:37:35 | preglow | dunno what it stands for, but it's essentially a dram design without the explicit use of a capacitor element for the the memory part |
15:37:49 | preglow | uses paracitic capacitance instead |
15:37:56 | preglow | which is rather clever :) |
15:38:21 | saratoga | I think its "1 tick" as in single cycle |
15:38:28 | preglow | hmm, nice |
15:38:39 | funman | saratoga: the AS3525 datasheet mentions 1T DRAM |
15:38:49 | preglow | never seen it myself, just briefly heard of it |
15:39:05 | funman | just buy a Sansa Clip to see it in action ;) |
15:39:28 | preglow | JdGordon: nah, this is lower level than i've been in the pcm stuff |
15:39:31 | preglow | JdGordon: what is it for? |
15:39:32 | * | linuxstb points to wikipedia http://en.wikipedia.org/wiki/Dynamic_random_access_memory#1T_DRAM |
15:39:42 | saratoga | hopefully the 1 T DRAM is the same as SRAM for our purposes |
15:40:09 | JdGordon | preglow: nuts :p well the idea was to be able to mix seperate pcm buffers as late as possible to get around issues like no voice while paused |
15:40:23 | saratoga | having all this IRAM would be very nice |
15:40:42 | saratoga | i think the awful performance per clock on the gigabeat F is due to lack of IRAM vs. PP |
15:41:02 | saratoga | if so, the AMS chips would be much faster then the Gigabeat F in most codecs |
15:41:53 | amiconn | The gigabeat F/X has similar performance per clock as the PP chips, as it's (only) armv4t as well. |
15:42:46 | amiconn | That's why the beast can handle ape -c4000 w/o problems at 264MHz, while the gigabeat F can't, even at its higher clok |
15:43:12 | amiconn | Oh, and on the F/X we don't use the available IRAM at all... |
15:44:20 | saratoga | amiconn: i remember looking at AAC/Ogg performance a while back and being surprised how much better PP did |
15:44:51 | saratoga | I think the Ogg needed something like 60MHz per 1x of realtime, vs. <40 for PP |
15:45:05 | saratoga | and this is inspite of the ARM9 load/store improvements |
15:46:06 | saratoga | hmm the 65MHz speed on the AMS's IRAM means its probably not single cycle then |
15:46:08 | | Join kugel [0] (n=chatzill@unaffiliated/kugel) |
15:46:36 | kugel | hello, can anyone tell me a approximate precision of the stopwatch plugin? |
15:47:31 | JdGordon | run it next to a real stopwatch for a while and see what happens? |
15:47:39 | funman | saratoga: look at page 23 of the as3525 datasheet |
15:48:10 | saratoga | oh "1T" == "1 transistor" |
15:48:16 | funman | 'one idle cycle for refresh is needed every 32 clock cycles' |
15:48:24 | GodEater | fixed.cfg is where you should put settings that you always want to remain the same on boot up right ? |
15:48:34 | saratoga | funman: thanks |
15:49:11 | amiconn | kugel: It uses the timer tick, so if timers are implemented correctly on the target, it's typical xtal precision, +/- 0.01s |
15:49:40 | funman | saratoga: where did you buy your JTAG connection ? I think I'll get one (if I get it too late for development, at least I'll use it to unbrick my other Clip) |
15:50:18 | kugel | Are they on a e200? I want to use it for an experiment and need to know the precision for the error analysis |
15:50:18 | moos | GodEater: yup |
15:50:46 | GodEater | so does fixed.cfg get completely wiped if you use the "reset settings" feature of the bootloaders ? |
15:50:53 | saratoga | funman: i picked some random ebay shop, and they shipped it from china |
15:50:53 | GodEater | or just ignored ? |
15:51:12 | JdGordon | GodEater: fixed.cfg should never be written to by rockbox |
15:51:13 | preglow | JdGordon: well, good luck on that, it sounds like something we want |
15:51:13 | saratoga | but theres a few places on google for about 10 USD |
15:51:34 | GodEater | JdGordon: so if someone does a settings reset, will it then be ignored that time ? |
15:51:37 | | Join HBK- [0] (i=hbk@pool-71-96-74-73.dfw.dsl-w.verizon.net) |
15:51:40 | GodEater | or will Rockbox still read from it ? |
15:51:40 | JdGordon | preglow: thanks :) but im so far out of my depth with that i wouldnt hold your breath :p |
15:51:49 | preglow | JdGordon: but how to handle dsp when? will it only be applied to the main pcm buffer? |
15:51:59 | preglow | as in the one that does music |
15:52:06 | preglow | s/when/then/ |
15:52:10 | | Quit HBK (Read error: 104 (Connection reset by peer)) |
15:52:11 | JdGordon | GodEater: I think fixed.cfg is only read at boot |
15:52:14 | | Quit {phoenix} ("Konversation terminated!") |
15:52:30 | GodEater | JdGordon: that's my point, when you do a settings reset, that's done at boot too |
15:52:45 | GodEater | so I'm wondering if the fixed.cfg could cause issues if someone forgets they've made one |
15:53:09 | amiconn | kugel: Timers are implemented correctly on PP, so there's just the +/-0.01s quantization error, plus the clock deviation which depends on the individual unit and is also temperature dependent. |
15:53:20 | JdGordon | preglow: the dsp is only used for music currently isnt it? does it make sense to put the voice audio through it? |
15:53:25 | funman | http://cgi.ebay.com/JTAG-Programmer-Adapter-Test-Debugger-PCB-JTAG20-Cable_W0QQitemZ350115670034QQcmdZViewItem?hash=item350115670034&_trksid=p3286.c0.m14&_trkparms=72%3A1240|66%3A3|65%3A12|39%3A2|240%3A1318 looks legit |
15:53:33 | linuxstb | GodEater: "reset settings on boot" isn't a bootloader feature - it's done in Rockbox itself. |
15:53:40 | amiconn | Typical error is a few seconds per day, like any digital watch/ clock |
15:53:42 | preglow | JdGordon: i don't know, really, i kind of thought it was applied globally |
15:53:49 | JdGordon | GodEater: umm yeah, if they reset at boot it could, I thought you are talking about resetting from the menu |
15:53:58 | preglow | JdGordon: well, it depends, would you expect the voice ui to get more treble if you adjusted the treble setting? |
15:54:12 | preglow | JdGordon: i kind of would expect something like that to apply globally |
15:54:43 | linuxstb | Isn't DSP needed for resampling voice? |
15:54:48 | preglow | that too |
15:54:52 | preglow | hmm, yeah |
15:54:56 | preglow | there's different dsp for voice now |
15:54:59 | preglow | remember now :> |
15:55:18 | * | GodEater makes a note of this info in his "possible gotchas" list |
15:55:22 | amiconn | preglow: Imo some dsp settings should apply globally, but not all (e.g. channel configuration - e.g. voice would be muted when karaoke would be applied to it) |
15:55:31 | preglow | amiconn: very good point |
15:55:35 | JdGordon | like I said... im way out of my depths here.. I figure if I get a base working either someone can help out, or take over... |
15:55:54 | preglow | JdGordon: jhMikeS has/had plans for a very nice mixing system |
15:56:02 | preglow | that would also solve the current issue we have with dsp latency |
15:56:08 | preglow | but it would require some fairly nice changes... |
15:56:20 | linuxstb | preglow: Can you remember what his plan was? |
15:56:50 | preglow | only that it involved moving dsp very late in the chain and almost certainly would require preemptive multitasking, tbanks to the latency :) |
15:57:04 | JdGordon | I couldnt find the discussion, but from what i remmeber im sort of close to it? |
15:57:51 | amiconn | preglow: I don't think we need preemptive multitasking. It would open a whole can of worms for almost no benefit. Just mix in an isr, using small blocks |
15:57:56 | jhMikeS | amiconn: you panged? |
15:58:07 | preglow | amiconn: might work, yes |
15:58:19 | preglow | but i'm still not convinced preemptive multitasking would be bad for rockbox |
15:58:35 | JdGordon | GodEater: it that something which we should fix? I cant image fixed.cfg is actually used by many people and especially not to set dangerous settings |
15:58:36 | * | jhMikeS hears "preemtive" and his rabbit ears perk up |
15:58:46 | preglow | :D |
15:59:27 | amiconn | jhMikeS: Yes. I have a strange mpeg2 video, which triggers 2 bugs in mpegplayer, a visual bug and a freeze bug |
15:59:52 | jhMikeS | amiconn: hmmm...unusual in which way? |
15:59:52 | saratoga | funman: check the shipping label, and thats where i got mine |
16:00 |
16:00:00 | preglow | it would be a very fun experiment anyway, one i hope jhMikeS will embark on one day :P |
16:00:04 | saratoga | funman: I mean to say that I checked my label and its the same |
16:00:25 | amiconn | This video has a max timestamp of 0xffffffff, but its min timestamp is > 0 |
16:00:39 | amiconn | So the difference is smaller than 0xffffffff, i.e. not invalid |
16:01:22 | amiconn | This causes the total playtime to be displayed as 26h 30min nn sec, which is too long on the X5 with 12-Nimbus and overlaps the icon in the center |
16:01:27 | GodEater | JdGordon: I don't know, it only just occured to me |
16:01:51 | amiconn | It also causes a hang with the disk spinning if you seek past the actual end of the video (video is ~15 min) |
16:02:04 | jhMikeS | amiconn: how long is the video in reality? |
16:02:11 | JdGordon | Am I wrong in thinking that if the DAC is set to 44.1khz then it will keep hitting the isr to try and stay at that rate regardless of how small the buffer it gets each time is? |
16:02:13 | jhMikeS | heh |
16:02:37 | preglow | i think the isr always assumes it gets what it asks for |
16:03:02 | preglow | at the lowest level, of course... |
16:03:25 | | Quit Darksair ("Reboot") |
16:03:32 | JdGordon | because (At least in the sim) if i fill 32k sample buffers it seems to work about right, but if that drops to 2 or 8 k samples it goes completly bad |
16:03:45 | jhMikeS | amiconn: I know min timestamps are quite often > 0. Things should be handled as unsigned. |
16:04:10 | amiconn | jhMikeS: Yes, and they are. Problem is that the max timestamp doesn't match reality |
16:04:18 | linuxstb | Aren't mpeg timestamps something unhelpful like 33-bit? |
16:04:20 | amiconn | ...and that causes trouble in mpegplayer |
16:04:37 | jhMikeS | linuxstb: yeah, but we use 45KHz and so ignore the LSb |
16:04:44 | funman | saratoga: thanks. how long did you wait for the shipping to arrive? |
16:05:23 | jhMikeS | amiconn: so the final stamp of 0xffffffff is actually valid for the file? |
16:05:25 | amiconn | Windows media player also has trouble playing that video. The file was produce by nero vision, and I'm sure there are no bit errors in it |
16:05:40 | amiconn | jhMikeS: no |
16:06:00 | amiconn | It should be much lower. I guess 0xffffffff actually means invalid |
16:06:02 | *** | Saving seen data "./dancer.seen" |
16:06:05 | amiconn | (or unknown) |
16:06:11 | JdGordon | GodEater: I cant ever remember anyone wondering why settings are not resetting problerly (because of fixed.cfg) but its a easy fix if it does come up |
16:07:10 | jhMikeS | amiconn: 0xffffffff does mean invalid in context though it is technically legal |
16:07:30 | | Join Darksair [0] (n=user@221.221.165.108) |
16:09:25 | jhMikeS | amiconn: what I'm asking is does the file actually contain a timestamp of 0xffffffff or is MPEG player failing to find the correct ending timestamp? |
16:10:17 | pondlife | JdGordon: A proper PCM mixer is a feature I'd like ;) Have you any design notes up? |
16:10:28 | jhMikeS | I guess running the SIM might give more info from the debug output |
16:11:49 | JdGordon | pondlife: design notes... hehehe.. uh, I mean, na, I figured id start by experimenting, then doing it properly |
16:12:05 | amiconn | jhMikeS: How could I find out? |
16:12:28 | amiconn | The sim says 0xffffffff (just in decimal) |
16:12:36 | preglow | JdGordon: a fine development method i tend to follow myself :P |
16:12:42 | pondlife | It's sometimes good to start with a list of requirements at least - once you get to that point.... and then an API for the new PCM layer... |
16:12:51 | JdGordon | pondlife: in all honesty I didnt design anything propeerly because I didnt eben know if I could get this far |
16:12:58 | * | jhMikeS checks what output he put in the sim again |
16:13:13 | pondlife | I'm very happy someone's getting on with it, so don't let me put you off :) |
16:13:38 | amiconn | jhMikeS: tons.... more than the standard windows command window backlog buffer |
16:14:08 | * | JdGordon thinks we should get pcm, playback and playlist out of apps and really fix up their interdependancies with the rest of apps |
16:14:23 | preglow | pcm, playback, def |
16:14:25 | preglow | playlist, i dunno |
16:14:35 | * | pondlife thinks we should scrap "apps" and have a set of layers |
16:14:47 | preglow | well, the current way is just two layers |
16:14:54 | preglow | nothing fancier than that |
16:14:55 | jhMikeS | amiconn: tons is true enough. but in the correct place for this problem? |
16:14:55 | pondlife | Yes, I want more! |
16:14:59 | preglow | pondlife: why? |
16:15:07 | preglow | pondlife: this would require a design document :) |
16:15:12 | pondlife | Of course! |
16:15:22 | jhMikeS | it should show result for each stream so the offending one can be id'ed. |
16:15:22 | JdGordon | more is good... the gui should have very specific privelages to real code |
16:16:09 | pondlife | We could do with more modularity, really. Currently we have just C function scope holding APIs together, might be useful to make it more explicit. |
16:16:14 | * | linuxstb doesn't think we need layers - just more independence between the different parts of the apps code |
16:16:26 | pondlife | Yes, that would be nice |
16:16:28 | * | linuxstb wants a whiteboard to scribble pictures onto |
16:16:41 | JdGordon | someone needs to invent multiplayer paint |
16:16:49 | JdGordon | multiplayer notepad just isnt good enough |
16:16:57 | * | pondlife wants a wiki page with the subsystems drawn up in a block diagram |
16:17:12 | preglow | go xfig! |
16:17:17 | * | linuxstb thinks we've strayed off-topic... |
16:17:29 | | Quit saratoga ("CGI:IRC (Ping timeout)") |
16:17:44 | | Quit homielowe () |
16:21:31 | * | linuxstb wonders what the 1.4MB of RAM usage in a typical swcodec build includes |
16:21:52 | JdGordon | the wps buffer would be a big part of that |
16:23:11 | JdGordon | 154880 bytes for the 220x176x16 lcd |
16:23:13 | | Quit Darksair ("To Arch or Gentoo? That is the question...") |
16:23:31 | JdGordon | so maybe not really big.. |
16:24:56 | jhMikeS | amiconn: can't the debug output be dumper to a file? |
16:25:05 | JdGordon | too many threads with largish stacks? is there a reason tagcache, dircache and playlist all need their own threads? |
16:25:15 | pondlife | linuxstb: Where do you get that 1.4MB figure from? |
16:25:28 | linuxstb | rockbox-info.txt |
16:26:11 | preglow | JdGordon: stuffing all of those in the same thread wouldn't be very wise, the thread separation probably helps avoid bugs |
16:26:13 | jhMikeS | JdGordon: tagcache, dircache obviously do since they also have background building |
16:26:23 | preglow | helps with abstraction |
16:26:31 | * | linuxstb spots 72KB of bss for icon.o |
16:26:37 | preglow | wtf |
16:26:44 | | Join Darksair [0] (n=user@221.221.165.108) |
16:26:54 | JdGordon | image buffer? |
16:27:32 | amiconn | jhMikeS: http://pastebin.ca/1237850 |
16:28:12 | JdGordon | yeah, icons do 24*24*bitdepth/8 * icon count (22 iirc) |
16:28:20 | pondlife | static unsigned char viewer_icon_buffer[NB_SCREENS][IMG_BUFSIZE]; |
16:28:23 | linuxstb | About 60KB for font.o, 64KB for unicode.o... |
16:29:06 | | Join PaulJam [0] (i=PaulJam_@vpn-3073.gwdg.de) |
16:29:13 | JdGordon | the global_settings sturct is nearly 1k! |
16:29:29 | | Quit kugel (Read error: 113 (No route to host)) |
16:29:36 | linuxstb | 40KB for dir_uncached.o, 35KB for dircache.o... |
16:30:16 | JdGordon | while this is fun.. is there actually anything that can be done about it? (without obviously dumping the feature) |
16:30:49 | pondlife | Could icon_buffer and viewer_icon_buffer be merged? |
16:30:55 | linuxstb | I don't know if anything can be done until we look at all the RAM usage... |
16:31:16 | * | jhMikeS wonders what's up with pastebin.ca ... main nav loads but paste isn't |
16:31:22 | | Join goffa [0] (n=goffa@216.220.23.105) |
16:31:41 | jhMikeS | meh, there it went |
16:32:12 | | Quit stoffel_ (Read error: 113 (No route to host)) |
16:34:06 | | Join J-23 [0] (n=kvirc@a105.net128.okay.pl) |
16:34:23 | | Quit Thundercloud (Remote closed the connection) |
16:34:48 | | Quit LambdaCalculus37 ("Ka-chunka") |
16:34:57 | JdGordon | tagcache.o and tagtree,o are the 2 biggest (by a decent margin) in apps/ |
16:35:31 | * | JdGordon finds he exacgurates more after 2am than usual :p |
16:36:38 | jhMikeS | amiconn: it looks like it didn't find any video timestamps |
16:37:10 | jhMikeS | just the first one anyway |
16:38:47 | funman | mpeg-2 stream? |
16:39:07 | jhMikeS | it does 1 & 2. they're aren't very different |
16:39:49 | | Join jhulst [0] (n=jhulst@unaffiliated/jhulst) |
16:39:56 | * | linuxstb wonders why the tagcache thread needs a 17KB stack |
16:41:26 | JdGordon | without ramcache my beast is using 27% of that 17k |
16:42:21 | JdGordon | if dircache cant be enabled at runtime without a reboot, why isnt its stack's buffer bffer_alloc()'ed like the rest of its needed buffers? |
16:43:56 | | Quit goffa_ (Read error: 110 (Connection timed out)) |
16:44:57 | * | jhMikeS wonders what the heck Nero did and whether it was compliant |
16:46:27 | funman | they say vlc's mpeg muxer and demuxer are a good reference (or were when they were written by Christophe Massiot) |
16:47:19 | funman | pretty much all video software authors add their own mess in the formats they write |
16:47:20 | jhMikeS | I do have all the ISO docs for reference |
16:47:52 | | Quit EspeonEefi ("ăăăȘă") |
16:47:54 | jhMikeS | though perhaps it has little heuristics that help |
16:47:56 | funman | if you can handle these docs they are better of course |
16:48:47 | preglow | JdGordon: if the thread isn't called at all if it's disabled, then it can be bufalloced |
16:49:22 | jhMikeS | they're the reason anything got done on mpegplayer in the first place. MPEG just makes you mash through the file and do your best to find things like the duration. |
16:49:58 | funman | there is also the difference between Program streams and Transport streams |
16:50:12 | linuxstb | IIUC, mpeg is just a stream of packets - there's no concept of duration, and timestamps can be discontinuous. |
16:52:29 | JdGordon | yeah, doesnt look like any reason to even create the dircache thread if its disabled |
16:53:29 | jhMikeS | fonman: we do only support transport streams here. |
16:53:57 | | Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr) |
16:53:58 | jhMikeS | those require PTS every 0.7s and it appears like Nero did something weird |
16:54:49 | funman | interestingly, vlc has an option to distrust timestamps and calculate them from the bitrate |
16:55:04 | funman | jhMikeS: ah ok |
16:55:15 | linuxstb | jhMikeS: Don't you mean Program Streams? |
16:55:44 | jhMikeS | isn't a program stream the container for muxing many channels for broadcast and video conferencing? |
16:56:01 | * | jhMikeS could have inverted the two |
16:56:04 | funman | I know transport stream can do that (it's designed for transport over a network) |
16:56:11 | funman | program stream is designed for storage |
16:56:40 | linuxstb | Transport Streams split streams into 188-byte packets - DVB is based on them. Program Streams are larger packets - DVD uses program streams with 2048 byte packets |
16:58:15 | | Join stoffel_ [0] (n=sfr@p57B4E8C6.dip.t-dialin.net) |
17:00 |
17:01:25 | funman | hum ts can mean Transport Stream as well as TimeStamp - reading this source is dangerous |
17:02:37 | jhMikeS | linuxstb: I guess my recollection just inverted the program/transport. |
17:03:18 | linuxstb | funman: I never see "ts" by itself (it doesn't make sense) - there is "dts" (decoder timestamp) and "pts" (presentation timestamp). But we're starting to stray from Rockbox.... |
17:03:36 | funman | mpeg_parser.c computes the packet length from the stream, so it's a PS |
17:03:58 | * | JdGordon upload patch FS #9506 and goes to bed |
17:04:00 | | Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon) |
17:04:35 | funman | linuxstb: see validate_timestamp() in mpeg_parser.c - I don't know if it's from rockbox or libmpeg2 though |
17:05:32 | linuxstb | Well, in that context it could make sense... |
17:05:35 | | Quit scorche (Nick collision from services.) |
17:06:29 | jhMikeS | it's from rockbox to make sure at timestamp is within the movie duration |
17:06:42 | | Join scorche [50] (n=scorche@rockbox/administrator/scorche) |
17:07:03 | funman | http://git.videolan.org/?p=vlc.git;a=blob;f=modules/demux/ps.c;h=274dc0bfee9c7f24a516e6b3e2a667aab26b0e16;hb=HEAD#l43 < the comment is not very helpful |
17:07:05 | jhMikeS | it's cruft anyway (there's a bit of that) |
17:13:56 | J-23 | can Rockbox play AAC/AAC+? |
17:15:23 | linuxstb | J-23: Yes to AAC. I _think_ the faster targets can play AAC+ |
17:15:52 | funman | AAC+ is High Efficienct profile? |
17:16:48 | | Join mf0102 [0] (n=michi@e181136069.adsl.alicedsl.de) |
17:18:24 | J-23 | funman: yes |
17:21:54 | | Join robin0800 [0] (n=robin080@cpc2-brig8-0-0-cust394.brig.cable.ntl.com) |
17:29:37 | | Join LambdaCalculus37 [0] (n=LambdaCa@c-24-0-218-198.hsd1.nj.comcast.net) |
17:33:49 | | Join nuonguy [0] (n=john@c-71-198-1-139.hsd1.ca.comcast.net) |
17:35:39 | | Quit AhtiK (Read error: 110 (Connection timed out)) |
17:39:23 | funman | is rockbox 3.0.1 still to be released ? |
17:39:26 | | Quit LambdaCalculus37 ("Ka-chunka") |
17:39:42 | linuxstb | It has been released. |
17:39:45 | jhMikeS | funman: after I fix the red on the table it's coming right out :) |
17:40:05 | funman | I wanted to correct http://www.rockbox.org/twiki/bin/view/Main/GeneralFAQ#What_is_Rockbox_What_is_its_purp |
17:41:10 | funman | but the main page still shows 'rockbox 3.0 released' in the Project news |
17:43:26 | | Join bmbl [0] (n=Miranda@unaffiliated/bmbl) |
17:43:30 | n1s | funman: 3.0.1 was only for archos targets |
17:44:20 | funman | isn't it worth mentioning anyway? |
17:44:30 | | Quit nuonguy ("Leaving") |
17:46:27 | | Quit Nico_P (Remote closed the connection) |
17:46:59 | | Join DerDome [0] (n=DerDome@dslb-082-083-225-159.pools.arcor-ip.net) |
17:50:57 | | Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky) |
17:52:07 | | Nick Darksair is now known as Darksair-{away} (n=user@221.221.165.108) |
17:53:33 | funman | Is it possible to easily fix the syntax of http://www.rockbox.org/twiki/bin/view/Main/SansaV2Firmware#Interrupt_Handlers ? |
17:54:12 | | Part pondlife |
17:55:59 | | Quit robin0800 (Remote closed the connection) |
17:57:06 | | Quit BigBambi ("http://www.mibbit.com ajax IRC Client") |
17:58:48 | n1s | jhMikeS, amiconn: do you have any thoughts on frequency scaling on the gigabeast? Should we go with a full-blown dynamic, interrupt-driven frequency/voltage scaling scheme or the reglar boost/unboost with 2 corresponding voltages? |
17:58:58 | | Join {phoenix} [0] (n=dirk@p54B47698.dip.t-dialin.net) |
18:00 |
18:03:19 | jhMikeS | n1s: I've thought it's worth investigating the full-blown stuff, at least the automated one. |
18:03:43 | domonoky | n1s: does the gigabeast have a seperate timer for the tick ? because on hxx0 there is the problem, that timers are not exact while switching frequencys, so we should switch seldom on these devices... |
18:04:09 | n1s | domonoky: nt a clue |
18:04:16 | n1s | s/nt/not/ |
18:04:19 | jhMikeS | domonoky: I think there are ways of latching for DVFS |
18:04:36 | domonoky | and the boost/unboost mechanism is already there, full blown frequency changing might require more changes to rockbox |
18:04:58 | jhMikeS | I was thinking CPUFREQ_NORMAL would be variable, so nothing radical would need doing |
18:05:40 | domonoky | and from what does the switching code know how high CPUFREQ_NORMAl should be ? |
18:05:58 | n1s | jhMikeS: btw, what's the word on the charging code? |
18:05:58 | domonoky | i think it surely would need input from either apps, or from the sheduler.. |
18:06:05 | *** | Saving seen data "./dancer.seen" |
18:06:13 | jhMikeS | the internal DVFS hardware will tell it by interrupt based upon hardware utilization |
18:07:12 | domonoky | ah, so the hardware knows how much it sleeps ? and if it doesnt sleep in a certain intervall, we would set the freq higher ? |
18:07:14 | jhMikeS | you have an option to implement your own algorithm or use the on-chip one |
18:07:51 | jhMikeS | it's based on that, cache usage, port activity and a number of other things. |
18:08:08 | jhMikeS | the details were somewhere or another online |
18:08:27 | * | domonoky doesnt know this special hardware on the beast, but if its possible without to much changes, it may be worth a try.. :-) |
18:08:44 | * | n1s reads data sheets |
18:09:14 | jhMikeS | one thing I want is to use that annoyance called SDMA so we have actual DMA for audio or whatever else. |
18:12:18 | | Join karashata [0] (n=kimi@69.41.192.215) |
18:12:47 | | Join LambdaCalculus37 [0] (n=LambdaCa@c-24-0-218-198.hsd1.nj.comcast.net) |
18:21:40 | n1s | wow, this thing is freakishly advanced: "The purpose of the DPTC module is to detect the minimum operation voltage for the IC, taking into account the extreme of processing activity and ambient temperature for a given frequency" ! |
18:23:33 | | Nick Darksair-{away} is now known as Darksair (n=user@221.221.165.108) |
18:23:34 | jhMikeS | the slightly less advanced version of this IC would have sufficed for a music player. I don't know what got up them. |
18:23:58 | domonoky | wow.. should allow some neat powersaving things :-) |
18:27:22 | | Quit stoffel_ (Read error: 113 (No route to host)) |
18:28:55 | | Join bughunter2 [0] (n=Jelle@77.164.66.126) |
18:29:16 | | Quit PaulJam (".") |
18:29:36 | | Join Strife89 [0] (n=michael@204.116.245.152) |
18:32:24 | | Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi) |
18:33:38 | | Join karashata_ [0] (n=kimi@69.41.192.215) |
18:33:48 | | Quit karashata (Read error: 113 (No route to host)) |
18:34:03 | | Nick karashata_ is now known as karashata (n=kimi@69.41.192.215) |
18:34:25 | | Quit jhMikeS (Nick collision from services.) |
18:34:31 | | Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS) |
18:36:58 | jhMikeS | is rockbox.org displaying strangely for anyone else, or is it my end? |
18:37:19 | LambdaCalculus37 | Looks fine here. |
18:37:55 | | Quit karashata (Read error: 104 (Connection reset by peer)) |
18:38:19 | | Join karashata [0] (n=kimi@69.41.192.215) |
18:38:31 | jhMikeS | lol. somehow I shut off my webfilter proxy by mistake. |
18:40:47 | bertrik | funman, a swiss company sells a nice not too expensive usb jtag adapter, they claim it works with openocd |
18:43:07 | funman | bertrik: too late I ordered the parallel adapter ;) |
18:43:28 | | Quit Darksair ("Zzz...") |
18:44:26 | jhMikeS | once again I've stuck the build server...strange spirits |
18:45:52 | | Join miepchen^schlaf [0] (n=miepchen@p579ECEF9.dip.t-dialin.net) |
18:46:31 | karashata | jhMikeS, I recall seeing an error about a type conflict between tick_funcs in kernel_sdl.c and kernel.h when I tried to compile the simulator |
18:46:58 | karashata | I changed the type in kernel_sdl.c to match kernel.h and it seems to compile fine now |
18:48:18 | jhMikeS | karashata: that should be fixed if you 'svn up' |
18:48:39 | | Quit bughunter2 ("bye") |
18:48:55 | funman | bertrik: I have some trouble playing with the peripheral clock frequency |
18:49:08 | karashata | ah, kay |
18:49:38 | funman | I tried to set it to the same frequency but using plla source rather than the 24MHz crystal directly - I get nothing on the screen anymore |
18:51:23 | | Quit Schmogel ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") |
18:51:40 | | Join pondlife [50] (n=Steve@rockbox/developer/pondlife) |
18:51:41 | bertrik | I haven't really looked into the clocks, but I think we should have a good look at it (/me invites mr someone) |
18:51:45 | | Quit pondlife ("Leaving.") |
18:52:10 | | Nick domonoky is now known as someone (n=Domonoky@rockbox/developer/domonoky) |
18:52:13 | funman | this afternoon I have found big parts of the code in OF which sets clocks frequencies / dividors |
18:52:24 | * | someone isnt good at this things :-) |
18:52:26 | funman | that is for all peripherals (nand, gpio, ide ..) |
18:52:28 | | Nick someone is now known as Domonoky (n=Domonoky@rockbox/developer/domonoky) |
18:53:01 | funman | I have not found evidence that the peripheral clock is based on something else than the 24MHz crystal |
18:55:17 | | Quit LambdaCalculus37 ("Ka-chunka") |
18:56:33 | | Join culture [0] (n=none@cpc1-bele3-0-0-cust658.belf.cable.ntl.com) |
18:56:43 | funman | I found 3 structures: 1 which holds 'desired' clock frequency for all devices, 1 which holds a couple of 'desired' PLL output frequency and its associated PLLA register setting; and one which holds device specific functions to set the clock divider |
18:57:20 | funman | that way we can know the operating frequency of each device |
19:00 |
19:00:40 | | Join stoffel_ [0] (n=sfr@p57B4E8C6.dip.t-dialin.net) |
19:01:11 | | Join neddy [0] (n=john@nat/sun/x-ade272801e3dee51) |
19:02:20 | | Join Aduch [0] (i=ravenz@i.hate.your.vhosts.shellium.org) |
19:02:30 | amiconn | n1s: It's probably worth checking out the dynamic method. As long as core voltage is fixed (Coldfire, PP, ..), frequency<->current relationship is (approximately) linear, but once dynamic voltage adjustment is added to the mix, the relation becomes (approximately) quadratic |
19:02:40 | bertrik | I think amiconn once told me that in PP devices, all clocks basically switch when boosting. It looks like we can avoid this on the as3525 |
19:03:08 | bertrik | i.e. keep peripheral clocks at the same frequency and boost only the cpu |
19:03:25 | n1s | amiconn: I'm reading the data sheet but my hope that i will be able to implement this is low. |
19:03:25 | | Quit jhulst (Read error: 110 (Connection timed out)) |
19:04:10 | bertrik | I think dynamic voltage adjustment is an advanced feature that should not be implemented in this early stage yet |
19:04:30 | | Join Nibbl [0] (n=Nibbler@port-83-236-56-2.dynamic.qsc.de) |
19:04:44 | amiconn | With that quadratic relation, it's better to dynamically adjust the clock than to switch back & forth between 2 fixed frequencies |
19:05:09 | bertrik | we should be able to do this on PP502x already... |
19:05:22 | amiconn | bertrik: We're talking about the beast here |
19:05:42 | amiconn | And we cannot do this properly on PP |
19:06:54 | amiconn | We do not know the core specs, and hence we have to assume that the required core voltage is fixed. |
19:07:55 | bertrik | ok, you mean it will involve guesswork to know what voltage is required at what frequency |
19:08:54 | amiconn | Yes, and there is no guarantee that a voltage that is sufficient for 30MHz on individual device X will be sufficient on individual device Y |
19:09:58 | amiconn | Even the fact that the voltage is changed on the fly could cause crashes or other erratic behaviour |
19:10:41 | amiconn | The beast is a completely different matter because we (a) have the datasheet and (b) this datasheet tells us that the core has dynamic voltage adjustment designed into it |
19:11:35 | bertrik | going back to the clip, I think a good first try would be to keep the peripherals at a fixed frequency (determined at compile time) and vary only the cpu frequency at run-time |
19:11:43 | | Part Aduch |
19:12:57 | bertrik | we'll need to know if any peripherals have any special requirements though (like certain maximum cpu/peripheral clock ratios etc.) |
19:14:04 | | Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net) |
19:18:13 | | Join mrkiko [0] (n=IRCExplo@151.80.154.87) |
19:25:45 | bertrik | funman, what pll divider values did you use? |
19:26:51 | | Join kachna [0] (n=kachna@r4ax178.net.upc.cz) |
19:29:34 | | Join Horscht [0] (n=Horscht@xbmc/user/horscht) |
19:32:54 | funman | 0x2630 = 384MHz |
19:33:26 | | Quit stoffel_ ("leaving") |
19:33:39 | funman | you can specify for each device if you use 24MHz crystal, plla, pllb, or an external clock (probably with an external oscillator) |
19:37:09 | bertrik | was it just the display or also other peripherals that stopped working (or even the cpu too)? |
19:37:34 | | Join jhulst [0] (n=jhulst@unaffiliated/jhulst) |
19:38:12 | * | bertrik tries to decode 0x2630 |
19:39:03 | funman | bertrik: look page 103 of the datasheet |
19:39:25 | funman | and page 105 for how to extract the 3 numbers from the 32 bits value |
19:40:10 | bertrik | I know, I have the datasheet |
19:40:19 | funman | ND = 2 (OD=01) , NF = 0x30 (48), NR = 6 |
19:42:09 | funman | if not modified, the lowest 7 bits of CGU_PERI are 0 (reference 24MHz, pclk 24MHz, extmem clock 24MHz) |
19:42:13 | | Join LambdaCalculus37 [0] (n=LambdaCa@c-24-0-218-198.hsd1.nj.comcast.net) |
19:43:03 | funman | I use CGU_PERI |= (0<<6)/*div0*/|(0xf<<2)/*div1*/|0x1/*plla*/; so pclk = 24MHz, and extmemclock = 384MHz (too much I know) |
19:43:03 | | Quit adamgolding (Connection timed out) |
19:43:27 | funman | but I expected the other peripherals (lcd) to function correctly |
19:47:54 | | Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net) |
19:48:30 | | Quit mrkiko ("leaving") |
19:49:44 | funman | 0x26C0 should have the same output |
19:49:58 | funman | 0x2C60* sorry |
19:50:08 | * | n1s makes the beast skip during playback :) |
19:50:43 | | Join Hillshum [0] (n=Hillshum@75-165-236-38.slkc.qwest.net) |
19:52:23 | n1s | well, at least setting different clocks works |
19:54:14 | bertrik | funman, sorry don't know what goes wrong. The datasheet mentions something about using a pre-defined order when switching clocks but doesn't further specify it. |
19:55:16 | funman | maybe to be sure to not exceed the mentioned requiremens |
19:55:33 | bertrik | Also, there may be a problem with synchronisation between the peripherals the cpu clock when they are not running from the exact same clock, so we probably need something other than the fastbus mode of the cpu. |
19:56:14 | funman | I don't know what's fastbus/asynchronous/synchronous mode :/ |
19:56:28 | funman | only saw it mentioned in the control register of cp15 |
19:59:30 | n1s | So, an initial experiment on the gigabeast yields some interesting results, writing to CLKCTL_MPCTL to change the clock seems to work, ticks seem to be directly coupled to this clock so if i lower the clock a tick takes proportionately longer time, if the disk spins down (and powers off) disk access will hang the system when the clock is lowered |
20:00 |
20:00:52 | Domonoky | n1s: so we would need to decouple the tick from the clock, either via a seoerate timer/counter if available, or via software.. |
20:01:20 | | Quit jhulst (Remote closed the connection) |
20:01:36 | n1s | Domonoky, it seems so, I haven't looked into how they are implemented yet |
20:02:28 | funman | bertrik: I can use that setup: PLLA = 0x2C10 (128MHz), cgu_peri: div0_sel=0, div1_sel=1 => extmem_clk = pclk = 64MHz |
20:06:06 | *** | Saving seen data "./dancer.seen" |
20:06:23 | bertrik | funman, I think I'll start looking at the sd interface again |
20:06:29 | funman | with IDE clock at 64MHz, MEMSTICK clock at 21MHz, still no luck for SD |
20:06:44 | funman | bertrik: ok, you mean reverse engineering the OF ? |
20:06:55 | bertrik | yes |
20:07:11 | | Join karashata_ [0] (n=kimi@69.41.192.215) |
20:07:15 | funman | you are looking at latest Clip Firmware, Vsomething.30 I think ? |
20:07:38 | bertrik | yes |
20:07:46 | | Quit karashata (Read error: 113 (No route to host)) |
20:07:56 | | Nick karashata_ is now known as karashata (n=kimi@69.41.192.215) |
20:08:08 | funman | I am still looking at the previous one, don't want to redo all the work; nor to see the bugs Sansa added since ;) |
20:09:58 | funman | at offset 0x6CD4 you have the operating clock frequencies of each device |
20:10:34 | funman | at 0x6D34 you have the couple {pll_output, pll_register_value} |
20:11:35 | bertrik | yuk, the sd code looks complicated |
20:11:49 | funman | yes :/ |
20:12:47 | funman | at 0x6E04 and 0x6E5C you have an array of SD-specific functions |
20:12:51 | funman | well, 2 arrays |
20:13:16 | funman | the first one is for NAND (through iNAND SD-NAND interface), the 2nd one is for SD (real SD for e200 for example) |
20:13:30 | funman | they call the same functions with a parameter (0 for NAND, 1 for SD) |
20:13:32 | bertrik | oh, the clip firmware contains e200 functionality? |
20:13:40 | funman | not really |
20:13:50 | funman | it contains useless functions however |
20:14:26 | funman | IIRC if you try to use the SD functions there is an error somewhere |
20:14:42 | funman | maybe their build system didn't permit them to remove these unused functions |
20:15:08 | funman | I mean they have no "#ifdef HAVE_SD_SLOT" |
20:15:29 | bertrik | all of the fancy compiler optimisation ruined by a stupid linker :/ |
20:15:33 | funman | ;) |
20:16:00 | funman | the first function is the "enable" one |
20:16:09 | funman | note the reference to '64MHz' |
20:17:11 | funman | you can also see the error 0xFFFFE403 being returned if you call this function with the 'SD slot' argument |
20:18:49 | funman | at 0x5962 we see loaded the NAF base address (usually used with the AMS NAND controller, but not here) for the NAND (0) case: look where it is stored then because the base register will not be explicitely loaded anymore in the OF ;) |
20:22:51 | | Quit karashata (Read error: 113 (No route to host)) |
20:22:53 | | Join karashata_ [0] (n=kimi@69.41.192.215) |
20:23:11 | | Nick karashata_ is now known as karashata (n=kimi@69.41.192.215) |
20:24:37 | | Join bluebrother [0] (n=dom@rockbox/staff/bluebrother) |
20:25:37 | | Join bughunter2 [0] (n=Jelle@77.164.66.126) |
20:26:11 | | Join Schmogel [0] (n=Miranda@p3EE21E75.dip0.t-ipconnect.de) |
20:38:55 | | Quit Strife89 (Nick collision from services.) |
20:40:06 | | Join ameyer [0] (n=ameyer17@adsl-75-57-173-227.dsl.emhril.sbcglobal.net) |
20:41:09 | funman | any e200v2 owner wanting to test some code? |
20:42:07 | Hillshum | what? |
20:42:48 | funman | just run what's in the git repository and see if the led blinks or not (after a ~20s delay) |
20:43:14 | Hillshum | give me a few minuets |
20:45:58 | | Join Hillshum_ [0] (n=chatzill@75-165-236-38.slkc.qwest.net) |
20:46:34 | Hillshum_ | funman: remind me where to start /rbutil? |
20:46:50 | funman | just run make in rbutil/mkamsboot/ once you built the bootloader |
20:47:37 | Hillshum_ | from where again? |
20:48:06 | funman | you must run tools/configure in any directory you like |
20:48:17 | Hillshum_ | oh yeah |
20:54:22 | Hillshum_ | funman: bootloader.bin? |
20:54:41 | funman | yep, BL=/blabla/bootlader.bin |
20:55:21 | Hillshum_ | i got an error 2 on the build |
20:56:43 | Hillshum_ | here's the output:http://rockbox.pastebin.com/f253ddf81 |
20:57:04 | funman | hm I thought there were placeholders |
20:57:14 | Hillshum_ | ? |
20:57:37 | | Quit crashmat1ix (Client Quit) |
20:57:48 | funman | let me fix that |
20:58:56 | bluebrother | hmm. This dlalyzer that got committed doesn't comply to the coding guidelines at all (typedef struct ...) |
20:59:36 | bluebrother | how are we dealing with such code? Is the rule more relaxed for utils? |
20:59:47 | funman | Hillshum_: fixed - you can git pull and re-run make |
20:59:54 | bluebrother | als, what do people think about caps in filenames? I personally dislike them ... |
21:00 |
21:05:52 | * | Domonoky thinks rules for tools are more relaxed.. :-) |
21:06:03 | Hillshum_ | funman ~35 secs and the wheel flashes |
21:06:52 | funman | \o/ |
21:07:01 | Hillshum_ | gotta go |
21:07:04 | | Quit Hillshum_ ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]") |
21:07:10 | linuxstb | funman: What was that? |
21:07:12 | funman | thanks ;) that means the SDRAM is access properly on e200 |
21:07:24 | linuxstb | Ah, that's good news. |
21:07:35 | linuxstb | It tested all 8MB? |
21:07:41 | funman | yes |
21:08:01 | funman | It needed this change: http://gitorious.org/projects/rockbox_sansa_v2/repos/mainline/commits/f549e9912ab38ba939ab0d301f61b4932c3ce86d - I still don't understand it |
21:08:06 | Hillshum | :D |
21:08:20 | funman | except that the value is 0x30000000 + 0x2300*number_of_MBytes |
21:08:34 | linuxstb | funman: Yes, I noticed that - it seems very odd |
21:09:33 | funman | some hidden 'feature' |
21:10:15 | Domonoky | looks strange.. reading from a specifc adress makes it work.. *scratches head* |
21:10:54 | funman | I remember all the RAM was filled with 0xff at init, I'll double check |
21:12:15 | Domonoky | maybe there is a self-clearing register there ? so reading could reset it to 0 causeing SDRam to work... *throwing wild guesses* |
21:12:42 | funman | do you mean self-refresh ? there is other means to control that |
21:14:19 | Domonoky | no, some microcontrollers have registers (normally some interrupt status or alike) which clears itself when you read it.. |
21:14:53 | funman | a register inside the addressable RAM ? :/ |
21:15:35 | Domonoky | memory mapped registers ? but it was just a wild guess... |
21:16:40 | | Join homielowe [0] (n=homielow@d206-116-134-81.bchsia.telus.net) |
21:17:32 | | Quit J-23 (Read error: 113 (No route to host)) |
21:17:33 | Domonoky | does the OF also read from this addresses in the init ? |
21:18:41 | | Join IudeX [0] (n=52a0f8f7@gateway/web/cgi-irc/labb.contactor.se/x-c06e251894869a14) |
21:18:45 | | Join fyre^OS [0] (n=nnscript@cpe-72-229-213-199.nyc.res.rr.com) |
21:18:53 | funman | yes - I didn't figure it by trial over the 500k+ addresses ;) |
21:19:08 | Domonoky | true :-) |
21:19:14 | IudeX | hey funman are u working on e200v2 or clip now? |
21:19:29 | funman | IudeX: Clip |
21:19:55 | funman | but sometimes I look at the e200 firmware and I figure the (very) tiny difference between the Clip and the e200 |
21:19:58 | IudeX | funman: on git i saw e200 commits. |
21:20:27 | IudeX | i'll try something today |
21:20:27 | Hillshum | I just tested them |
21:22:08 | bmbl | So what are the next major steps on the v2s? |
21:22:10 | | Quit IudeX (Client Quit) |
21:22:27 | Hillshum | SD and LCD |
21:22:43 | | Join IudeX [0] (n=52a0f8f7@gateway/web/cgi-irc/labb.contactor.se/x-b391a13d97cc1884) |
21:23:02 | funman | lcd is missing for e200, fuze, and c200 |
21:23:35 | funman | well what I read from that register is 0 .. interesting |
21:23:40 | | Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-1336fdc1754468eb) |
21:26:00 | | Join synergist [0] (i=christop@cant.be-arsed.co.uk) |
21:28:08 | | Join jhulst [0] (n=jhulst@unaffiliated/jhulst) |
21:30:12 | n1s | saratoga: It seems like the mdct codecs lib is built with the same O level as the core rockbox, is this intended? (might be worth checking out on Gigabeat at least) |
21:30:49 | | Quit karashata ("I go, only to return again some time...") |
21:31:17 | saratoga | n1s: which level is that? |
21:32:15 | n1s | saratoga: O1 for arm, Os for coldfire, IIRC i tried it with O2 on coldfire but it made no difference, and all arm targets _but_ the gigabeat use the asm mdct |
21:34:39 | | Quit fyrestorm (Read error: 110 (Connection timed out)) |
21:34:43 | saratoga | I need to investigate that further |
21:34:54 | saratoga | i'm skeptical that the c version should be faster on the gigabeat, and if it is, i would like to know why |
21:35:36 | | Quit IudeX ("CGI:IRC") |
21:49:59 | | Join adamgolding [0] (i=adamgold@i209-195-79-79.cia.com) |
21:59:25 | | Quit LambdaCalculus37 ("Ka-chunka") |
22:00 |
22:02:42 | | Join shotofadds [0] (n=rob@rockbox/developer/shotofadds) |
22:06:07 | *** | Saving seen data "./dancer.seen" |
22:10:23 | | Quit adamgolding (Read error: 110 (Connection timed out)) |
22:11:37 | | Join Seed [0] (n=ben@bzq-84-108-232-45.cablep.bezeqint.net) |
22:19:56 | | Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere) |
22:23:23 | | Join Thundercloud [0] (n=thunderc@cpc1-hem18-0-0-cust660.lutn.cable.ntl.com) |
22:28:50 | | Quit Nibbl ("Ex-Chat") |
22:35:27 | funman | mcuelenaere: what does "Yes (no rockbox driver yet)" mean for USB support in TargetStatus ? |
22:35:45 | mcuelenaere | you mean for the Onda targets? |
22:35:55 | funman | yes |
22:36:10 | mcuelenaere | it means: I got it working, but I haven't wrote a Rockbox compatible USB driver (yet) |
22:36:19 | mcuelenaere | it's the same story for the audio |
22:37:05 | | Quit neddy (Connection timed out) |
22:43:19 | | Part evilnick |
22:46:43 | | Quit mf0102 ("Ex-Chat") |
22:47:57 | | Join goffa_ [0] (n=goffa@216.220.23.105) |
22:49:04 | | Join PaulJam [0] (i=PaulJam_@vpn-3067.gwdg.de) |
22:56:43 | | Join kushal_12_27_200 [0] (n=kushal@12.169.180.178) |
22:57:09 | | Quit Schmogel ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") |
22:57:42 | | Quit Seed ("cu, Andre") |
22:59:32 | | Quit Domonoky (Read error: 104 (Connection reset by peer)) |
22:59:38 | | Quit goffa (Read error: 110 (Connection timed out)) |
23:00 |
23:00:35 | | Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]") |
23:04:12 | | Quit martian67 ("gone") |
23:08:25 | | Quit bmbl ("Woah!") |
23:08:42 | | Join deader [0] (n=deader@71-80-183-71.dhcp.azus.ca.charter.com) |
23:08:57 | | Join Zarggg [0] (n=zarggg@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com) |
23:08:57 | | Quit deader (Client Quit) |
23:09:53 | | Join dead3r [0] (n=dead3r@71-80-183-71.dhcp.azus.ca.charter.com) |
23:10:36 | dead3r | are the downloads not working for rockbox? |
23:10:52 | n1s | turns out that changing CLKCTL_MPCTL is probably not the way to go, as that is the clock for the whole SoC so a lot of different clocks are derived from it, there is a separate divider for the arm core though, but changing that makes the beast hang :( |
23:10:56 | dead3r | any release i download is just over 2.2 mb |
23:11:13 | BigBambi | dead3r: Which file? |
23:11:30 | dead3r | http://download.rockbox.org/release/3.0/rockbox-ipodvideo64mb-3.0.zip |
23:11:32 | | Quit DerDome ("Leaving.") |
23:11:52 | BigBambi | and what do you mean any release? |
23:12:00 | n1s | why is 2.2MB wrong? |
23:12:10 | BigBambi | That looks fine to me |
23:12:14 | dead3r | oh |
23:12:19 | dead3r | it's not unzipping |
23:12:20 | dead3r | then |
23:12:31 | n1s | dead3r: are you on a mac? |
23:12:40 | dead3r | yeah |
23:12:43 | gevaerts | unzips fine here |
23:12:47 | dead3r | do i need a 3rd party zip? |
23:13:00 | BigBambi | mac OS X is a bit braindead on unzipping |
23:13:11 | n1s | dead3r: the dir in the zip starts with a dot so the friendly unzipping app hides it :) |
23:13:11 | BigBambi | use rbutil, it is much easier |
23:13:18 | linuxstb | Not OS X, just Finder. The command-line "unzip" program works fine. |
23:13:35 | BigBambi | linuxstb: Sorry, yes |
23:14:41 | linuxstb | dead3r: We recommend this - http://www.rockbox.org/twiki/bin/view/Main/RockboxUtility |
23:15:24 | dead3r | thankyou sir |
23:16:13 | | Quit Thundercloud (Remote closed the connection) |
23:17:34 | dead3r | detected an unsupported apple player varient |
23:17:36 | dead3r | hm |
23:18:11 | | Quit Rob2222 (Read error: 104 (Connection reset by peer)) |
23:18:12 | BigBambi | Is it a classic? |
23:18:52 | | Join Rob2222 [0] (n=Miranda@p4FDCE7D9.dip.t-dialin.net) |
23:18:58 | dead3r | 80gig vid |
23:19:09 | dead3r | has cover flow |
23:19:13 | BigBambi | That is a classic |
23:19:23 | BigBambi | And Rockbox doesn't work on it |
23:19:32 | dead3r | wat |
23:19:33 | dead3r | :( |
23:19:43 | | Quit ender` (Read error: 104 (Connection reset by peer)) |
23:19:43 | dead3r | ok |
23:20:15 | | Join ender` [0] (i=krneki@foo.eternallybored.org) |
23:20:40 | | Quit havien (Remote closed the connection) |
23:20:54 | | Join havien [0] (n=none@68-189-143-101.dhcp.wlwl.wa.charter.com) |
23:21:06 | dead3r | i'm guessing the support for it is still in development? |
23:21:32 | scorche|sh | no one is working on it at the moment |
23:21:34 | BigBambi | Not really |
23:21:48 | BigBambi | dead3r: It is encrypted and new undocumented hardware |
23:22:30 | BigBambi | Baring a major unexpected and remarkably unlikely breakthrough, you won't see it for a long long long long time, if ever |
23:22:34 | dead3r | oh i see |
23:22:39 | BigBambi | My money is on never personally |
23:23:08 | dead3r | brings to mind the PSP |
23:23:18 | dead3r | too bad |
23:25:14 | * | bluebrother likes the "unsupported player variant" detection :) |
23:25:37 | linuxstb | bluebrother: Doesn't rbutil say what it's detected? |
23:26:50 | bluebrother | it says that it detected a player, but the detected player is unsupported |
23:27:48 | dead3r | i think mine's a 6th gen, no? |
23:28:09 | bluebrother | hmm, how are stacks handled in the sim? It seems that trying to solve an empty sudoku hangs in an endless (?) loop until stkov |
23:28:14 | bluebrother | but in the sim it works fine. |
23:28:34 | BigBambi | dead3r: yes, that is what a classic is |
23:29:17 | funman | bluebrother: It probably uses the host' stack |
23:29:28 | linuxstb | It does. |
23:29:29 | funman | not sure if it's given by linker or operating system |
23:29:52 | bluebrother | anyone ever tried using the same stack on the sim? |
23:30:13 | funman | bluebrother: try "ulimit -s yourstacklimit" |
23:32:12 | | Join lwizardl [0] (n=jamesdav@76-221-209-202.lightspeed.gdrpmi.sbcglobal.net) |
23:32:13 | lwizardl | hi |
23:32:53 | lwizardl | i'm having trouble installing rockbox on my black 30gb ipod video. using the windows installer |
23:32:58 | | Quit kushal_12_27_200 ("Leaving") |
23:33:42 | bluebrother | so ... how does this trouble look like? |
23:34:10 | lwizardl | in the installer it fails the complete install |
23:34:48 | | Quit bertrik ("Leaving") |
23:34:56 | bluebrother | and I bet it fails with *drumrolls* an error message! |
23:35:09 | lwizardl | yeah |
23:35:25 | bluebrother | so, you're expecting me to guess your error message? |
23:35:38 | lwizardl | failed to install bootloader |
23:36:03 | bluebrother | ok. How did you configure? Did you use the autodetection button? |
23:36:38 | lwizardl | yes it detected the player but not the type |
23:36:55 | bluebrother | the player but not the type? |
23:37:05 | bluebrother | so what did it detect? |
23:37:13 | lwizardl | it found its mount point, and the opened the apple tab but didn't highlight the model |
23:37:27 | | Quit n1s () |
23:37:35 | bluebrother | did you do a manual install / a previous install? |
23:37:40 | lwizardl | no |
23:38:21 | funman | can you try on a USB1.1 connection ? |
23:38:36 | lwizardl | all my ports on the machine are 2.0 |
23:38:36 | bluebrother | do you have a hub between PC and player? |
23:38:52 | lwizardl | bluebrother: nope direct to the ports on the back of the pc |
23:38:52 | | Quit {phoenix} (Remote closed the connection) |
23:39:06 | bluebrother | can you try a different cable? |
23:39:17 | lwizardl | nope only cable i have |
23:39:57 | bluebrother | well, it looks like ipodpatcher isn't recognizing the Ipod. There have been reports for this happening due to bad cables or usb ports. |
23:40:00 | funman | say 'abracadabra' and try again ;) |
23:40:13 | lwizardl | trying different ports |
23:40:26 | | Quit petur ("*plop*") |
23:40:27 | bluebrother | so you could try to use a different usb port. If that fails try to borrow a cable somewhere |
23:40:38 | bluebrother | or try a different computer. |
23:40:39 | lwizardl | that worked |
23:40:52 | lwizardl | port next to the nic cable worked |
23:41:02 | lwizardl | installing fonts now |
23:41:36 | | Quit mcuelenaere () |
23:43:12 | | Quit dead3r () |
23:45:44 | lwizardl | ok it said i have to manually install the themes |
23:48:59 | lwizardl | but i'm not seeing the instructions on how to do it. can someone point me in the direction |
23:49:44 | bluebrother | http://www.rockbox.org/wiki/WpsGallery |
23:51:09 | lwizardl | yeah i'm looking at that page. when i select my ipod version it just takes me to a place to download various themes |
23:51:27 | lwizardl | but does not explain on how to install them |
23:53:14 | bluebrother | just unzip the files to the player |
23:53:30 | lwizardl | to the root? |
23:54:06 | bluebrother | and "rockbox theme install" is a good term for asking google ... |
23:54:17 | bluebrother | yes, to the root. If that doesn't work the theme is not packaged properly |
23:55:09 | | Quit moos ("Rockbox rules the DAP world") |
23:55:46 | lwizardl | ok |