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 2005-05-09

00:01:17preglowstuff from 2003 is very unlikely to even patch cleanly
00:02:04Bagdert0mas: here?
00:07:03 Join LinusN [0] (
00:07:46BagderI don't see how bmp.c can work on the sim as it is now
00:07:52Bagderhence the warnings
00:08:18Bagderso my fix didn't actually add the warnings
00:08:23Bagderit just made them visible
00:09:29HClyou can definately close the create playlist with search one
00:09:33amiconnBagder: Yes, I also found that
00:09:40HCldatabox+searchengine should surpass that by farrrrr
00:10:24preglowHCl: done
00:11:20amiconnThe H-140 on ebay today went up to >400 :(
00:11:36HClwow o.o
00:11:44HCli can sell my player with a profit!
00:12:48preglowwhat's the difference between closed and deleted, really?
00:13:11ashridaharghl. wrong channel
00:13:23Bagderpreglow: I guess its just a visible difference
00:13:28Bagderand for statistics etc
00:13:32 Part Moof
00:13:34amiconnA joke:
00:13:50preglowBagder: so should i delete or close these old patches?
00:14:01Bagderwe always use close on them
00:16:25HClwe have rockblox patches in the patch section
00:16:27HClfor iriver :3
00:16:40HClwe should test those
00:21:25 Quit ashridah ("Leaving")
00:23:53 Nick ac_away is now known as austriancoder (~austrianc@
00:25:14*preglow gets bitten by his first 64 bit incompatible code
00:26:15Bagder64 bits add some fun effects at times if you haven't been careful
00:28:16Bagderthere - back to the usual build-yellows now
00:28:29preglowfun sound effects in this particular instance
00:29:46preglowthis is of course not my code
00:30:12***Saving seen data "./dancer.seen"
00:30:22 Quit Sucka ("a bird in the bush is worth two in your house")
00:30:30preglowi make far more stupid bugs than 64 bit mixups
00:30:39Bagderwell, I still occationally fall over 64 bit problems so I have sympathy ;-)
00:30:50Bagderin my own code I mean
00:31:16 Quit arfo ("see ya")
00:31:35preglowi've grown quite fond of size_t and those boys after getting a 64 bit cpu
00:31:43preglow64 bits of precision in ints = sweet
00:31:48Bagdertime_t then?
00:32:09Bagderit isn't 64 bit on all 64 bit oses
00:32:16Bagdersome have it 32
00:32:35preglowthat's quite brain dead
00:32:43amiconn...and on some systems, size_t is 64 bits even on a 32 bit cpu
00:32:59preglowlet's hope they fix that in about fourty years
00:33:00amiconn...e.g. cygwin, I had to learn that :(
00:33:15Bagderamiconn: you sure about that? aren't you thinking about off_t ?
00:33:30amiconnMaybe that, yeps
00:33:49Bagderoff_t is 64bit on all modern systems
00:34:09Bagderexcept windows
00:34:29Bagderon windows, off_t is 32 bit
00:34:41Bagderif you don't use cygwin that is
00:34:52Bagderwell, msvc and borland at least
00:35:19HClif i had a sourceforge account
00:35:21amiconnoff_t doesn't sound like a windows api thing. Pure windows api needs to support 64 bit offsets
00:35:33Bagderoff_t is for posix
00:35:46HClyou could put me down for 675003, 887324, 562120...
00:35:46preglowmy 2001fp is a couple of days away!!!
00:35:54 Join bagawk [0] (~Lee@bagawk.user)
00:37:01preglowhcl: i can't find any of those...
00:37:06HClthey're under requests
00:37:20preglowahh, requests
00:38:09preglowholy sheeyat, the requests section sorely needs some janitoring
00:38:31Bagderwe're drowning
00:38:45austriancoderwhats the max len of o line code in rockbox?
00:39:00Bagderor less than 80 even
00:39:09austriancoderok.. i have on with 108.. will change it
00:43:22t0mas[00:02:14] <Bagder> t0mas: here? <−− yes
00:43:48Bagdert0mas: check my bmp.c commit
00:43:56Bagderyou never tried that code in a sim, did you?
00:44:18t0masI did
00:44:19t0masIriver sim
00:45:03Bagdervery strange
00:45:09 Quit ]RowaN[ ()
00:45:54t0mas+#define readshort(x) *(x)
00:45:54t0mas+#define readlong(x) *(x)
00:46:06Bagdersince you pass in pointers
00:46:09t0mashm... sounds logic to me
00:46:18 Nick Aditya|GYM is now known as Aditya (
00:46:19t0masbut it worked w.o it afaik
00:46:23Adityabe back.. booting into linux
00:46:32Adityaerr hmm
00:46:35Adityajust realized..
00:46:38AdityaI dont have linux..
00:46:47Adityawell.. going to install ..
00:46:52Adityabe back
00:47:00t0masok, that's a nice one...
00:47:05amiconnLinusN: I've updated my 'unused code' table according to today's changes
00:47:11 Quit Aditya (Read error: 104 (Connection reset by peer))
00:47:29*LinusN hasn't installed openoffice on his linux machine
00:47:51t0masthat sounds better
00:48:04t0masBagder? have you tested the "wrong" code?
00:48:19Bagderbut we got 5 warnings
00:48:21t0mashm.. I'm pretty sure it worked...
00:48:33t0masbecause I tested it on sim...
00:48:48t0masdebugf()-ing out the width, height and bytecont and things like that
00:49:57amiconnLinusN: The html export does look a bit odd, but anyway:">
00:54:34*t0mas slaps bagawk for taking the Ba[tab]
00:54:45t0masBagder: It's compiling now :)
00:54:52t0masI'll test tomorrow morning...
00:55:08bagawkt0mas, ?
00:55:24t0masyour nick is just before Bagder in the tab list ;)
00:55:40 Nick Bagder is now known as Baagder (
00:55:48 Nick bagawk is now known as _bagawk (~Lee@bagawk.user)
00:55:51t0masthat's better
00:55:58t0mastnx both :)
00:56:01 Nick Baagder is now known as Bagder (
00:56:33t0masIf it works... I'll scream for help to find out how that's possible ok?
00:57:51Bagderfair enough!
01:01:20t0masgood night :)
01:02:12preglowi am so intensely happy i'll never know how many hours i've spent in hunting for the most trivial of bugs
01:02:46*HCl likes trivial bugs.
01:03:25*t0mas remembers a discussion on a mailinglist a few weeks ago...
01:03:34preglowafter having spent well nigh three hours hunting for them, i like them to be spectacular
01:03:44t0masa file edited by at least 8 people... all done things to the same part
01:03:50t0masand in the end there's a bug...
01:04:09*t0mas did cvs annotate, some name came up for that line
01:04:14HClpreglow: i guess...
01:04:22HCli guess its just that i once had a bug
01:04:27t0masand with that single email... I started a 30 mails long discussion about who broke it :P
01:04:31HClof which 3 people searched for during 2 months
01:04:35HCland we all got so tired of not finding it
01:04:41HClthat we dropped the entire projcet
01:05:02HClt0mas: rather than discussing who broke it, they should fix it.
01:05:11t0masyes, it was already fixed
01:05:19t0masit was really trivial
01:05:47t0masbut the one who wrote that line didn't break it... he said... as someone else did something....... (you get the point)
01:07:40 Join Aditya [0] (
01:08:12_bagawkHCl is bad for you (the chemical formula that is)
01:08:29Adityawell.. I forgot I had taken out my second HDD and put it on the server.. which is of course not working.. oh well.. I guess I will just leave it off till next weekend.. or something
01:10:21t0masin #ipodlinux there is some guy NaCl
01:10:27t0masthat's not to bad for you :)
01:10:27HCli saw.
01:10:31HClcommon house salt.
01:10:39*HCl munches on NH4Cl
01:10:41t0masloser :P
01:10:49HCl :p
01:11:05*HCl is looking at the rio riot playlist generation features
01:11:09preglowwhat, we've all got evil twins in #ipodlinux?
01:11:23*preglow goes looking for postglow
01:11:28HCli think i'll add an "random" option to the searchengine soon
01:11:37HClmaybe more...
01:11:40HClpreglow: :P
01:12:18HCltoo bad iriver doesn't have an rtc...
01:12:36Bagderindeed a silly decision
01:12:55HCli'd bluntly say it was a bad one
01:13:30HCli'm looking at the "rio dj" features.. i'm wanting most of them into rockbox
01:14:04t0mashm... what do you need an rtc for? playing wakeup music in the morning?
01:14:11t0masoh wait...
01:14:13*LinusN is so tired of all the clueless people who think a good pseudo random generator requires an rtc
01:14:18HClkeeping track of time. its handy for all sorts of statistics.
01:14:22t0maslast played in the past ... hourd
01:14:30HClyea, for example.
01:15:00t0masLinusN: nobody mentioned a pseudo random generator
01:15:07LinusN"it is impossible to get good random on the H120 because it lacks an rtc"
01:15:16Bagdert0mas: not now, but lots have done before
01:15:25t0mas:| damn...
01:15:29LinusNa common misticriver topic
01:15:39t0masqbasic programmers :P
01:15:43t0masrandomize timer
01:15:47preglownot programmers at all, actually
01:15:49t0masnumber = 100 * RND
01:16:03t0masw/o even declaring number as an int or whatever :)
01:16:31preglowdon't diss dynamic typing, that'll end in a duel
01:16:54*HCl goes to copy the rio karma features to the wiki and write down whats needed before they're possible...
01:16:56t0mashaha, ok... I just have some bad memorys on basic...
01:17:04t0mashad to learn it on school...
01:17:08preglowbasic was my first language
01:17:09t0masvisual basic
01:17:13t0masmine too :)
01:17:19t0masQuick basic... when I was 12
01:17:32Bagderbasic v2 (in the C64)!
01:17:43preglowstarted on the c64 myself
01:17:47t0masyeah, but I'm not that old ;)
01:17:48preglowbut didn't quite get what was going on then
01:17:52amiconnsinclair basic here
01:18:44t0masbut after that came Visual Basic... in school...
01:19:50HCli'm wondering here
01:19:56HClif i copy a feature list of rio karma
01:20:01HCland list whats needed to implement that in rockbox
01:20:06HClis that breaking any copyright anywhere?
01:21:04t0mashm... technically... maybe?
01:21:06Adityapretty sure its not
01:21:09t0masas they wrote it
01:21:12t0masand you republish it
01:21:17Adityaits only what the rio karma does
01:21:18t0masbut if it's just a simple list...
01:21:20HClwell, its not their code, just their idea.
01:21:20Adityalots of places do it
01:21:33t0masHCl: code and docs or whatever.... it's all text...
01:21:58t0mascopyright laws don't care what it is... just who wrote it and what you're doing with it.
01:22:32t0maslike the white hairy animal? or are you going to bed? ;)
01:23:33_bagawkMy teacher at school still uses quickbasic to teach basic programming class...
01:24:28t0mashmm... there is something to say for that...
01:25:04t0masas in visual basic people will start clicking around and paint an interface... instead of thinking how to program should work...
01:25:24t0masand there is no logic flow thgough it... it all gets called on events...
01:25:42_bagawkSeems like a good okay language to start out some programming concepts on, but that program is getting _old_ and barely runs on a NT based OS
01:26:13t0masthere are ports...
01:26:50Adityawhats wrong with events?
01:26:58AdityaOOP is crazy useful
01:27:45t0masyes it is
01:27:52t0masbut when learning how to code...
01:28:44preglowbut no
01:28:45preglowtime to bed
01:28:50t0masDon't know exactly why... but it feels logic to start in a "line by line" language... so you can follow what it does... and not have something jump arount events
01:28:53t0masnight preglow
01:28:54 Quit preglow ("leaving")
01:29:43CoCoLUSi think students should start with c
01:29:48CoCoLUSthen progress to c++
01:29:57CoCoLUSbecause c is like... the mother of all languages :)
01:30:13_bagawkNo need to learn one then go to another...
01:30:31_bagawkJust makes you erase some things you already learned
01:30:39CoCoLUSwell c++ introduces some oop features...
01:30:47t0masCoCoLUS: but c in school is to difficult...
01:30:53t0masphp is better then
01:31:04CoCoLUSthats an entirely different world if you ask me
01:31:08t0masimagine for example my former school class..
01:31:30t0mas24 kids.... 2 or 3 can code basic... the other 21 just can't code anything...
01:31:41CoCoLUSwell it depends on the "school" i guess...
01:31:48t0masthats true
01:31:49*HCl adds to wiki
01:31:50HCllooks good.
01:32:02CoCoLUSsome "i design my own webpage" 2 hours a week course or some really computer focused school
01:32:07HClout of the 10 playing modes of the rio karma
01:32:17t0masCoCoLUS: the first one would be to difficult :P
01:32:18HCl5 are green
01:32:27t0masthe "1 hour per week learning colors in word" school
01:32:30zethe karma's got 10 playing modes?
01:32:32HCland 4 are blue (possible, but not yet)
01:32:37HCland 1 is red
01:32:45zeHCl: url?
01:32:46HClze: well, the "Rio DJ" has 10 playing modes
01:32:53HClfor the original manual
01:33:01HClfor the wiki
01:33:54zei almost never use the rio dj myself heh
01:34:06HClyou have one?
01:34:15HCli see
01:34:19HClwhats it like?
01:34:24zeits nice
01:34:27HCli considered it a lot before i decided on iriver
01:34:32HClmostly due to its rio dj
01:34:37zecould be nicer with better firmware, but the fw's perfectly fine as it is
01:34:50CoCoLUSunlike the iriver players...
01:35:01HClbut eventually the hdd size and lack of radio made me decide for iriver
01:35:02CoCoLUSgapless playback *cough*
01:35:09DMJChey... I like my iriver
01:35:24HCloh. right.
01:35:24zeheh yeah its gapless to the extreme... even crossfade's if you want
01:35:36CoCoLUSshut up :P
01:35:44DMJCbtw how do you update the original firmware once rockbox is on?
01:35:58zebut its WPS isn't customizable
01:36:01CoCoLUSyou don't ?
01:36:09HClheh o.o.
01:36:12CoCoLUSwhy update the original if you have rockbox up and running?
01:36:23zeand it seems physically capable of more than it does
01:36:23DMJCbecause rockbox can't do anything yet
01:36:37CoCoLUSah you're speaking of the actual now ;)
01:36:41DMJCI mean... it makes a nice game machine
01:36:42*HCl just thought up a new ability for the searchengine
01:36:53DMJCbut it's not exactly playing music anytime soon
01:36:55HCl"and random %30 of the time"
01:36:57t0masDMJC: if anything usefull comes out from iriver... I guess LinusN will update the bootloader..
01:37:02CoCoLUSyeah but it works perfectly
01:37:15t0massome time
01:37:40t0masbut I don't expect iriver to come up with something we really want...
01:37:44CoCoLUSpatch new firmware with bootloader -> upload that image with the original firmware -> good to go
01:37:47zeone thing that bugs me about the karma's firmware is browsing by genre throws away any artist/album hierarchy and just shows loose tracks
01:37:59 Part MoosCamaro
01:38:03HClze: there's simply no other way..
01:38:04amiconnHCl: Didn't you want to start on the runtimedb? ;)
01:38:10HClamiconn: yes
01:38:22HClamiconn: i'm mostly waiting for someone *cough* linus *cough* to get mp3 playback to work.
01:38:31HClits hard for me to develop without having a working copy on my player
01:38:32CoCoLUSwhere is he, anyway?
01:38:36HCli can barely use sims
01:38:40zeHCl: i think 100% of my albums have self-consistent genre tagging
01:38:52HClze: not all people though.
01:38:52t0masCoCoLUS: he is here... and he was on vacation
01:38:58CoCoLUSi see
01:39:13CoCoLUSa relaxing one, i hope
01:39:17HClrockbox supports a search for genre at the moment
01:39:17zei can never fit much of it into any official genre's anyway, so i mostly use it for broad categorization, and having the hierarchy would be useful to me with that
01:39:27zewhereas the loose tracks as it stands is simply worthless
01:39:27HClbut no browsing
01:39:41*HCl read the manual
01:39:46HCli figured that was worthless.
01:40:00HClthe problem is that albums can have songs with lots of different genres.
01:40:10zeyeah but thats only for VA albums generally
01:40:20AdityaHcl.. implement a keyboard? that would be pretty neat..
01:40:22*amiconn seldom uses genre tagging
01:40:23zeand while some people have a lot of those, a lot of people don't really
01:40:23HClyou'd prolly have to use something like the code i have with −−dirisalbumname in
01:40:29HClAditya: we already have one.
01:40:54amiconnIt's so vague... my genre taginfo is practically always empty
01:40:57Adityaso uhhh.. just have a search box that searches all the info about a song.. thats possible no?
01:40:57zemy genre's are usually 'odd' 'obscured' 'fucked' etc
01:41:20HClAditya: define "searching all the info about a song"
01:41:39Adityasearches song title, artist, album, genre, filename
01:41:47HClwe have that.
01:41:55HCldatabox+searchengine can do that no problem :)
01:41:58Adityaerr.. maybe I missed something in the combo
01:42:07HClit can also do things like..
01:42:17HClall songs where title == albumname
01:42:18HCletc :P
01:42:43HClits not done yet...
01:42:46Adityaget that bootloader going! I want this shit on my H320 :P
01:43:01*LinusN looks away
01:43:09Adityaheh j/k
01:43:09DMJCbootloader already works
01:43:15Adityafor H3xx?
01:43:18zehow about an evolutionary generative music maker?
01:43:46zeit makes stuff up as it goes and you have controls to say "i like what you're doing now" or "i hate that" or "that 0wnz me silly"
01:43:58*HCl grins.
01:44:02zeand after a while it gets to where it 0wnz you silly all the time
01:44:03HClthat would be sick.
01:44:13HCli actually know someone who made software that did just that.
01:44:22HClmaybe i should ask him about it.
01:44:27DMJCheh but all it'd spit out is metal music..
01:44:28Adityalots of people are working on stuff like that
01:44:45HClyea, it had an interface with simply a sad smiley and a good smiley
01:44:52HCland eventually, if you stayed at it long enough
01:44:56HClit figured out your music profile
01:45:08HClmmm, i remember that inputting and describing what each song was like was a pain though
01:45:12HClokay, bad idea.
01:45:20HClwe don't want to have to describe each song in detail to a computer
01:45:23HClin order to enable that :P
01:45:50HClcause you need to put in values like "loud" "aggressive" "soft" etc.
01:46:20*HCl goes some brainstorming about what would be needed in the searchengine to enable things..
01:46:36HClmaybe someone can help me think a bit here :P
01:46:50HClpretty much, i have an array that contains 0 and 1's, and i know the index of each
01:46:56HClif its a 1, its a possible candidate
01:47:08HClhow do i properly limit the selection to a certain amount of time
01:47:19HClwithout losing its randomness?
01:47:28HCllike, playtime in total
01:48:37CoCoLUSi don't need software to figure out my taste
01:48:42CoCoLUSi -know- what i like listening to?
01:49:12HCl :P
01:49:25HClyea, i just scrapped the idea on the grounds that its too much trouble describing each song anyways
01:49:27CoCoLUSon top of that, it simply is impossible
01:49:35CoCoLUShow should it choose music?
01:49:44HClit actually did a fairly decent job.
01:49:58CoCoLUSlike "oh your skin temperature is like 2 degrees low today, you sure are in a DEPRESSIVE mood"
01:50:06CoCoLUSwhen it was just raining...
01:50:08HClyou say "like" or "dislike"
01:50:14HCland it'll adjust to that.
01:50:25HClfind me a solution for the problem i described :/
01:51:06CoCoLUSmissing sound playback? abduct linus ;)
01:52:05*HCl forces a smile, best he can think of so far is making a temporary array in which you store the indexes of positions that have 1's, then using a random number to select songs from that till you have them all or till you reached your time quota..
01:52:11t0mashm... poor Linus... he shouldn't had told everybody about the 5 secs of music...
01:53:14*HCl goes to work on his ebnf for the searchengine...
01:53:16DMJCwouldn't it be easier to make an app that mixed existing songs
01:53:21DMJCto get the one you prefer
01:54:03_bagawkWTF! opera 8 as just shown it's first bug to me
01:54:21_bagawkIt is using all my memory
01:54:29_bagawk500MB pagefile now being used
01:54:34_bagawk(I have 256 mb ram)
01:54:46_bagawkANd it is going up
01:55:12HCl :p
01:55:22_bagawkI have never had a virus
01:56:50austriancoderhas somebady a nice sample.wav to test audio output?
01:58:16_bagawkaustriancoder, make one yourself?
01:58:20thegeek[01:56:32] [_bagawk] I have never had a virus
01:58:49thegeekhighly improbable
01:58:53_bagawkthegeek, as far as I know it is true...
01:59:09_bagawkI keep my machine clean
01:59:13austriancoderahhh.. audacity seems to be fine
02:00:16_bagawkMy parents have had, which is why I bought my own HD and setup a dual boot (and because they would not let me install anything, they thougt i would get a virus :P)
02:04:50*austriancoder has broken i2c on his iriver..
02:06:04*HCl yawns
02:06:12HClany requests for search engine capabilities?
02:06:58*HCl is extending the language with a few things like title startswith night, artist ends with jackson, all songs year >=1980 and random 30% of the time
02:07:20HClfilter playtimelimit 15min..
02:08:57amiconnHow do you mean, random?
02:09:07amiconnHow can a search be random ????
02:09:19HCllike it'll do a random 3 that can be 1,2 or 3
02:09:23HCland you check for 1 all the time
02:09:30HClallowing you to select a random 30%
02:09:32HClof the current search
02:10:48*HCl needs a good way to limit the accumulated playtime of the results...
02:11:22Adityado the rating thing WMP does
02:11:28HClrating thing?
02:11:29Adityaits actually pretty smart
02:11:41Adityait relatively rates songs based on how many times ur have listened to them
02:11:48HCloh. that.
02:11:59HClthats already in the design
02:12:04HClas well as a user based rating
02:12:07Adityaplaycount is useless
02:12:11Adityaafter like 1000
02:12:32AdityaI mean like.. if a song has 1000 listens.. and another has 2000, the first would have 1 star and the second would have 2 starts
02:12:39Adityaso its always relative
02:12:56HClthats linear...
02:13:07HClyou can just search for all songs between 1500 and 2500 listens to get 2 star songs
02:13:56AdityaI dont think I am getting my idea across
02:14:03Adityahypothetical example:
02:14:07AdityaI have 5 songs
02:14:19Adityathey have the following playcounts
02:14:52Adityason g#5 would have 5 stars
02:15:15Adityathe others would have [their playcount]/10000 stars (rounded of course)
02:15:30HCl1000, i assume
02:15:33HClwhat about 5000 ?
02:15:33Adityaerr.. ([their playcount]/10000) * 5
02:15:55Adityawhat about 5000?
02:16:08HCl10000 is a set thing or just the maximum of the whole?
02:16:21Adityaits just the maximum out of the current playcounts
02:16:27HClfair enough
02:16:33HClshould be almost trivial to implement.
02:16:37HCli'll add it
02:16:39Adityaits basically like a graph
02:17:01Adityak cool
02:17:15Adityaor you could be fancy and say smartRaTiNg
02:17:19HClthat just means its in the language, not that its supported
02:17:23Adityabecause you know.. bullshit like that is all the fad now :P
02:17:26HCli don't call such a rating very smart ;p
02:17:54HCli prefer auto cause it sortof implies its automatically generated and likely to have flaws.
02:18:00Adityayeah.. I was j/k
02:18:04HCl :P
02:18:14Adityait doesn't really have flaws.. its actually pretty useful
02:18:23Adityaas apposed to raw playcounts
02:19:12*HCl decides the filter command should be a sort of "and"
02:19:22*HCl has it almost worked out too.
02:20:10HClmeh, just trying to think of ways to implement the new language bits...
02:21:17Adityaman.. studying sucks
02:21:19Adityaits so boring
02:22:39 Nick StrathAFK is now known as Strath (
02:23:25 Part LinusN
02:25:37austriancoderhow can i set a 8th bit of regs[O_MBDR]?
02:27:22HClregs[O_MBDR]|=0x128; ?
02:27:29HClwait o.o
02:27:32HCli mean
02:27:34HCl0x80 or 128
02:27:36HClnot 0x128
02:28:00austriancoderwhat?! +g+
02:28:31HCli dunno, it was just my uneducated guess, knowing that thats how you set a bit in an int
02:28:35HCli dunno about regs :3
02:28:51*austriancoder will do some tests with the old i2c code
02:29:10*HCl will go to sleep.
02:29:15HClbusy day tomorrow...
02:30:14***Saving seen data "./dancer.seen"
02:30:52austriancoderdoes chaning volume in audiotest realy works?
02:31:14HClno idea
02:31:55austriancoderwill run a last check tonight.. need some sleep
02:37:44 Quit austriancoder ("using sirc version 2.211+KSIRC/1.3.12")
02:52:27 Join ehntoo [0] (
02:57:33 Quit _bagawk ("Leaving")
03:12:35 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
03:18:45 Join StrathAFK [0] (
03:33:20 Quit Strath (Read error: 110 (Connection timed out))
04:02:10 Quit QT (Read error: 131 (Connection reset by peer))
04:05:29 Join QT [0] (as@area51.users.madwifi)
04:30:16***Saving seen data "./dancer.seen"
04:38:57 Join |it|ChasKi` [0] (
05:03:40 Quit StrathAFK (Read error: 110 (Connection timed out))
05:12:13 Quit asdsd____ (Read error: 60 (Operation timed out))
05:23:03 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
06:30:20***Saving seen data "./dancer.seen"
06:34:01 Join Harpy [0] (
06:40:47 Join |it|ChasKi` [0] (
06:49:59 Join Strath [0] (
06:52:36 Join matsl [0] (
07:12:02 Quit Rick (Read error: 131 (Connection reset by peer))
07:12:15 Join Gibbed [0] (
07:20:11 Quit |it|ChasKi` (Read error: 145 (Connection timed out))
07:20:35 Quit matsl (Remote closed the connection)
07:25:05 Quit Stryke` ("Friends don't let friends listen to Anti-Flag")
07:50:35 Join ashridah [0] (
07:51:07 Quit Aditya ("Download Gaim:")
08:00:41 Join StrathAFK [0] (
08:18:18 Quit Strath (Read error: 110 (Connection timed out))
08:30:21***Saving seen data "./dancer.seen"
08:45:22 Join LinusN [0] (
08:51:49 Join dwihno [0] (~dw@
08:53:50 Join |it|ChasKi` [0] (
09:25:00 Join courtc_ [0] (
09:28:27 Quit courtc (Read error: 110 (Connection timed out))
09:31:39 Join bobTHC [0] (
09:31:49bobTHCmornin' folks !
09:33:50 Quit |it|ChasKi` (Read error: 60 (Operation timed out))
09:37:11 Join lImbus [0] (
09:38:31lImbushi all
09:38:46lImbusBagder, are you there ?
09:39:01lImbusor any other mailing list moderator ?
09:39:16LinusNi am here
09:40:25lImbusso how should I behave if I want to send in a new francais.lang ?
09:40:33LinusNuse the patch tracker
09:40:35lImbusmy mail to the list has been rejected my the mod
09:40:42LinusNi rejected it
09:40:59lImbusmy submissions have been forgotten more often in the patch tracker than on the list / irc...
09:41:25lImbusbut I will, itf it's the wanted procedure
09:41:49lImbusshould I post something somewhere if it's done ?
09:41:58LinusNwhat other patches from you were forgotten?
09:42:14LinusNlImbus: we get email confirmation of all patch entries
09:42:16lImbuslangstuff, included in the meantime
09:43:10t0mashm... if someone can explain me what I should check and how... then I can check lang files and commit them?
09:43:27t0masas I've heard a lot of them comming by the last weeks
09:44:43LinusNt0mas: most of the time you can just commit them
09:44:55t0masok, amiconn told me I should check something
09:59:51LinusNlImbus: your new-francais.lang file has changed the line trminations from LF to CR/LF
10:01:42LinusNlImbus: have you tried these strings in the trigger settings?
10:07:57 Quit Ka (Read error: 113 (No route to host))
10:12:17 Join Ka [0] (
10:18:07 Quit lImbus (Read error: 113 (No route to host))
10:19:13 Join lImbus [0] (
10:19:26lImbusSorry LinusN, my internet connection is going wild
10:19:31lImbusI am reading the logs
10:19:50t0masno need to lImbus
10:19:58lImbusI'll fix the line terminations
10:20:03lImbust0mas: aha ?
10:20:03t0masnothing was said :)
10:20:39lImbusyes, something was said, and my client didn't show it. my last line is 9:44 your post
10:20:55t0masoh ok, [10:18:17] * lImbus has quit (Read error: 113 (No route to host))
10:21:03t0masthat was the time you quitted :)
10:21:56t0masbut eh... LinusN already corrected it I guess
10:24:51lImbusdon't think so. looks like he's willing to teach (and force) me a bit today. He could have acceppted the submission to the list as well.
10:27:33 Join Lynx_ [0] (HydraIRC@
10:30:25***Saving seen data "./dancer.seen"
10:42:11LinusNlImbus: i have committed your patch
10:42:44LinusNand the reason we reject huge attachments on the mailing list is purely to spare the subscribers from large emails
10:43:15amiconnt0mas: Usually I check a few things before committing a language update. (1) Is is complete? No reason to not commit if not so, but I could then try contacting the submitter
10:43:52amiconn(2) Is it correct? I.e. all deprecated strings deprecated etc. Easy as simply running uplang on it and checking the output
10:44:25amiconn(3) There are some strings for which it is crucial that they're short enough
10:44:46lImbusLinusN: yes, I understand that. I only saw it was 60 KB after reading it's been rejected. However, I resisted to zip and resend it because any of the core developers/maillinglist-mods could see it.
10:45:11t0masok, so next time someone has a patch... and we desparate for developers here... I can run uplang, check line-ends and commit? :)
10:45:21t0mas*we are
10:45:32lImbusLinusN: I found a solution for the trigger strings. so I am going to update the patch soon
10:45:42LinusNlImbus: goodie
10:45:52LinusNt0mas: basically yes
10:47:20LinusNsure, you might screw up that particular language, but it can be fixed afterwards if that happens
10:47:45LinusNour lang patch maintainer, quelsaruk, is not around that much nowadays
10:48:25t0mashaha, confidence... :)
10:54:34LinusNanyone with an H3x0 here?
10:55:44 Join |it|ChasKi` [0] (
10:57:22 Join austriancoder [0] (~austrianc@
10:57:29austriancoderhi all
10:58:33 Quit Ka (Read error: 113 (No route to host))
11:02:17 Join amiconn_ [0] (
11:03:21 Nick austriancoder is now known as ac_learning (~austrianc@
11:03:46 Nick courtc_ is now known as courtc (
11:09:59 Join aliask [0] (
11:10:32aliaskLinusN: I have a H340.
11:12:57*rasher points to
11:14:10 Quit amiconn (Nick collision from services.)
11:14:11 Nick amiconn_ is now known as amiconn (
11:14:40aliaskWhen you say testing, what do you mean? Software only, or hardware hacks?
11:15:21rashersoftware, and I guess adding h340 is in order
11:15:42aliaskOk then. I just registered, so I'm editing it in now.
11:15:52rasherwell any testing really, it's not like this is a legally binding contract :)
11:18:14aliaskWell, I love my mp3 player a lot. So I dont really want to break it.
11:19:52 Join Ka [0] (
11:33:18LinusNaliask: does it come with a remote?
11:37:28ashridahthe 3xx comes with a remote, just not one with an LCD.
11:37:41ashridah(although it WORKS with one with an lcd)
11:42:43LinusNi don't have the lcd-less remote
11:42:54amiconnBagder: around?
11:42:57LinusNdoes it have the same buttons as the lcd remote?
11:51:50 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
11:55:24lImbusLinusN, thanks for francais.lang :-)
11:58:17 Quit DMJC (Read error: 104 (Connection reset by peer))
11:59:35 Join DMJC [0] (
12:01:14 Join MoosCamaro [0] (
12:01:38MoosCamaroHey all
12:07:28 Join preglow [0] (
12:08:17 Quit lImbus (" ooops")
12:15:23rasherafternoon, preglow
12:18:42aliaskLinusN: The LCD remote buttons are different. It has some for actually controling the unit if I remember correctly. (Up, Down, Enter etc)
12:18:54Bagderamiconn: here now
12:23:59 Quit DMJC (Read error: 104 (Connection reset by peer))
12:28:54amiconnBagder: I found what causes the strange 'rm' lines on 'make clean'; they're part of the dumb Makefile
12:29:25amiconnBtw, did you already check why the dependency of the binary on librockbox.a broke?
12:29:37Bagderno I didn't
12:30:15BagderI think I removed it while working with my fixes, and I need to put it back
12:30:28***Saving seen data "./dancer.seen"
12:30:48BagderI've made an effort to reduce the amount of sub-makes from within makefiles
12:37:01 Join asdsd____ [0] (
12:38:35 Part asdsd____
12:40:01 Join DMJC [0] (
12:40:26 Quit ehntoo (Read error: 104 (Connection reset by peer))
12:46:42 Quit ferenczy (Read error: 148 (No route to host))
13:14:55HClrasher: instead of actually merging searchengine with databox, i think i just want to do an plugin_exec on searchengine at the end of databox..
13:17:28HClto allow "builtin" search querys in the menu somewhere, much like rio karma
13:20:59 Join |it|ChasKi` [0] (
13:33:38*HCl yawns
13:48:27 Join tvelocity [0] (
13:52:34 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
13:54:42 Join Lost-ash [0] (
13:57:50ac_learningcan sombody test, if changing volume in audio test on iriver works, or not?
14:01:43 Quit tvelocity ()
14:02:01rasherac_learning: it doesn't change anything
14:02:24rasherinterestingly, it just crashed
14:04:34 Join b0bTHC [0] (
14:05:50ac_learningrasher: thanks..
14:12:26 Quit bobTHC (Nick collision from services.)
14:12:53 Nick b0bTHC is now known as bobTHC (
14:13:04 Quit DMJC ("Leaving")
14:16:50 Quit ashridah (Read error: 113 (No route to host))
14:17:57 Quit Seed (Nick collision from services.)
14:18:04 Join Seed [0] (
14:30:30***Saving seen data "./dancer.seen"
14:38:14 Quit Lynx_ (" reboot")
14:47:14 Nick Lost-ash is now known as ashridah (
14:50:24 Quit ashridah ("Leaving")
14:52:43preglowchaning volume works here?
14:53:13rasherdoes it affect the audio test though?
14:53:15preglowchanging, yes
14:53:17rasherI thin that was the question
14:53:26preglowwhat do you mean by that? yes, it alters the volume
14:53:40preglowyou have to keep pressing for a decade, though
14:54:05rasherlet me try again
14:54:19preglowand you have to actually press it
14:54:20preglownot hold it
14:54:54rashererp, keeps crashing
14:55:01rasherplays like half a second
14:55:06rasherthen hangs
14:55:59preglowworks nicely here
14:56:03preglowthe first time around, at least
14:56:34rasherthis is the second time it's crashed
14:56:42preglowrun it two times, and you're sure to have a crash
14:56:54rasherwithout me having touched anything after boting
14:58:22rasherand again
15:00:40rasherand again
15:00:43rasherwith another wav
15:02:30LinusNrasher: try increasing the disk spindown timeout
15:04:41preglowoh, so you haven't figured out that buyg
15:05:05rasherIndeed, that was my first thought, but I thought that had been fixed
15:05:14LinusNme too :-)
15:05:42LinusNhmmm, the code looks ok
15:06:24preglowLinusN: i think the entire wavpack codec will fit in iram ;)
15:06:48preglowit runs very well as it is
15:07:00rasherwhat the..
15:07:08rasherwhat should I set disk spindown to?
15:08:29rasherset it to 254 now, still it crashes after half a second
15:18:56LinusNis it any different if you run at 120MHz?
15:22:50 Join |it|ChasKi` [0] (
15:35:25rasherlet me see..
15:35:39aliaskNight all.
15:35:52 Quit aliask ("Chatzilla 0.9.67 [Firefox 1.0.3/20050414]")
15:36:08rasherthat works
15:36:44rasherthe last few seconds were static
15:36:59rasher"Catch the digital [NOISE]"
15:37:18preglowthat's just the automatic song improvement heuristics kicking in
15:37:38rasherdidn't crash though
15:38:15rasherstill crashes at 48mhz
15:41:20 Join ferenczy [0] (
15:54:07 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
16:12:04t0masat release time it would be nice to have some "rockbox song"
16:12:16Bagderfeel free to record one! ;-)
16:12:24*t0mas only plays guitar...
16:12:30t0masneed some more people then :)
16:12:50Bagderdon't look at me!
16:12:58*LinusN runs away
16:13:13t0masintro: "Catch the renewed digital flow! Oh, do I hear a gap? No, it's GAPLESS!|"
16:13:18t0masor something like that :P
16:13:18preglowsomeone make vocals for it
16:13:28preglowcatch the digital flow
16:13:51*ac_learning plays guitar
16:14:14*LinusN plays mp3
16:14:26preglowi demand this rockbox song be recorded with "little sound dj" in rockboy lists 13 songs named "rock box"
16:14:29rasherand c64! I have evidence
16:15:02bobTHCa song or just a "Radio commercial like "?
16:15:23t0masdoesn't matter to me
16:15:25preglowi don't care, as long as it contains arpeggios, pulse wave sweeps and filters
16:15:32t0masjust something to say like ":P" to iriver
16:16:26t0masbut we first need mp3 playback to listen to it ;) and gapless to make it fun ;)
16:16:44Bagderconvert one of Linus' c64 songs to mp3 :-)
16:16:56rasherI could have one ready in a matter of minutes
16:16:57t0maswhat was his nick in that time?
16:17:44Bagderallthemusic also lists one song named "rock boy" ;-)
16:17:45rasherif you have the HVSC it's under various/a-f/Boogaloo/
16:18:02preglowthat's cheap, someone complete a sid player
16:18:19rasheryou're our codec guy :(
16:18:37rashernot that that's bad, just that you're the only one :(
16:18:40preglowthe codec guy steers clear of that particular codec
16:19:06rasherSpeex appears to have fixed-point mode by the way
16:19:11preglowthough it's not that much of a problem
16:19:20preglowfind a fast 6510 emulator, and bolt in resid
16:19:29preglowit will never be cycle accurate, though
16:20:16rasherI don't think anyone would be expecting that
16:20:18Bagderit doesn't need to by cycle accurate
16:20:21t0maspreglow: have a good reason not to try sid?
16:20:32preglowt0mas: nah, it'll just need a lot of opts, i think
16:20:46preglowt0mas: and i've got lots of other things planned for when i have rockbox time again
16:21:14*t0mas starts thinking of a thing to do when he's helpdesking again this summer...
16:21:24LinusN"there's always time for rockbox"
16:21:39t0mas(everybody prepare for a big load of programming question this summer)
16:22:03preglowi'm done at uni around june/july, so i'll see then
16:22:37*t0mas has 2 months to fill... and only money for 2 weeks of vacation...
16:23:01preglowdrop the vacation and live like a count while you code rockbox instead
16:23:21t0maspreglow: you know how I code...
16:23:32preglowcan't say i do
16:23:44t0maswrite 2 lines... ask something... nobody answers... see it's trivial... write 2 more lines... ask again
16:23:55preglowi can relate to that...
16:24:17LinusNpreglow: still no clue regarding the imdct_l performance drop?
16:24:19t0masas I don't know that much rockbox internals...
16:24:34t0masand I never really learned C... started in C++ right away
16:24:38preglowLinusN: haven't had time to look at it
16:24:45LinusNme neither
16:24:48preglowLinusN: there's something funky happening before the loop
16:25:23preglowwell, yeah
16:25:30preglowi've tried exiting before the loop
16:25:35preglowthat should cut execution time lots
16:25:37preglowbut it doesn't
16:26:04preglowthe old imdct_l with c windowing is tons and tons faster then just the new imdct_l without the loop + windowing
16:26:10preglowand that's seriously, seriously wrong
16:26:32rasherIt's taunting you.
16:27:49t0masyeah... preglow can't rest before it works.. he atleast he knows why it doesn't work?
16:29:14preglowthat's not very far from the truth
16:30:34***Saving seen data "./dancer.seen"
16:30:48preglowi did spend quite some hours writing it, hoping that i'd at least get a minor speed improvement, so the least i'd want from it is learning why i should never try something like that again ;)
16:31:07t0masghehe ok
16:34:47LinusNpreglow: you just end the function right before the loop?
16:35:03t0maswhat's the simplest codec there is in rockbox?
16:35:08t0masso I have some example?
16:36:33preglowLinusN: deed
16:36:52preglowt0mas: conceptually or in code?
16:36:57t0masin code?
16:37:31preglowof the transform codecs, liba52 is simplest
16:37:49preglowof the lossless ones, wavpack
16:38:02preglowlibflac's code is an abomination unto god
16:38:55Bagderthe libflac code is... different ;-)
16:39:04t0mashm... and stevenm wrote midi as a plugin?
16:39:18preglowlibflac's code is one of the most extreme cases of "OO" c i've ever seen
16:39:36*LinusN gas to go
16:39:42LinusNbye all
16:39:47 Part LinusN
16:40:03Bagderand he uses function names that are looooooooooooooooong
16:40:48t0masah, I hate that...
16:40:53t0masit's more typing :P
16:41:52preglowof course they're long, he's simulating long chains of namespaceses, classes and subclasses
16:42:21preglowhe even goes as far as naming things 'private', 'public' and 'protected'
16:42:33BagderC++ damaged ;-)
16:42:48preglowi think c++ is ok, just not in the form of c code
16:44:46*t0mas likes c++ too...
16:44:54t0masall other things I code are c++
16:45:28BagderI'm generally not fond of C++
16:45:50Bagdereven if I do use it at times
16:46:06t0mashm... I like the classes and namespaces idea...
16:46:11preglowi like it better for larger things
16:46:15t0masas it keeps things separate..
16:46:18Bagdernamespaces are good
16:46:18preglowsince it tends to force me to think
16:46:36t0masAnd I agree with preglow... you have a better overview when things get large..
16:47:04Bagderunfortunately, people tend to over-c++ things that make things harder
16:47:22preglowyes, of course
16:47:29pregloweverything with a pinch of moderation
16:47:57Bagderyes, when you offer people a powerful tool, they make powerful errors ;-)
16:48:28t0masis the sid fileformat documented somewhere?
16:48:41Bagderits not a file format really
16:48:45Bagderyou need to emulate the CPU
16:48:49Bagderand the sound chip
16:49:08t0mashm... sounds slow..
16:49:11preglowa sid file is really just a program
16:49:11Bagdersid being the sound chip of the c64
16:49:46t0masand how should a codec in rockbox output?
16:49:47preglownot too slow
16:49:53preglowthe 6510 is a really simple processor
16:50:01preglowand it's not clocked very high
16:50:06Bagderlda #$00, sta $d020
16:50:22bobTHCooooooooo, so fast
16:50:24preglowindeed, i did some 6502 programming three-four years ago, and it's almost the same chip
16:50:30Bagderand 2-3 cycles per instruction
16:50:51t0masok, should be possible to emulate at a good speed..
16:50:59preglowyes, but you need the sid chip as well
16:51:07preglowand i recommend the resid library for that, as long as it's fast enough
16:51:09preglowit's pure fixed point
16:51:18t0masI know, just downleded it :)
16:51:24preglowif i don't remember incorrectly
16:52:37t0masbut it's c++
16:52:43t0masto that will be some porting work
16:52:45preglowthat's so
16:52:49preglowgcc does c++
16:52:56Bagderwe can deal with C++
16:52:58preglowyou just need to wrap it
16:53:04preglowit's not extensive c++
16:53:07preglowit just puts things in classes
16:53:24preglowjust compile it without exceptions and shit
16:53:45t0masok, so C++ isn't a problem?
16:53:55t0masthen I can just change as little as possible in the lib...
16:54:11preglowno, c++ should compile fine
16:54:14preglowlibrary usage is another thing
16:54:22t0masname mangeling...
16:54:22preglowbut i think resid is fairly clean
16:56:29t0mashm.... resid is kind of heavy...
16:56:38t0masmaybe use a less cpu eating lib? :)
16:57:09HClyou people should've told me that c++ was fine too.
16:57:33Bagderwell, fine for imports
16:57:41HClmrf :(
16:57:55HClwell, i still went a tiny bit OO with databox, using structs
16:57:56preglowt0mas: how do you know it's heavy?
16:58:06t0masthey say it in the readme file
16:58:29t0mas"To our knowledge reSID is by far the most accurate SID emulator ever created.
16:58:29t0masThis comes at a price; what is considered a fairly fast CPU at the time of
16:58:29t0masthis writing is needed to run the emulator."
16:58:42preglowi think it'll be able to run at decent speedfs
16:58:55preglowno oversampling, of course
17:00:57preglowyou could of course try to port the original sidplay
17:00:59preglowbut it sounds like shit
17:01:49t0mashm... I'll look at it when helpdesking... as there's nobody around then... and I have to sit there anyway
17:08:46preglowi think most aspects of resid are spot on
17:08:52preglowapart from the filter, which of course sucks
17:09:24HCli need to talk to someone who knows about plugin loading...
17:09:45*HCl goes to clean his room...
17:12:15 Join webguest26 [0] (
17:12:26Bagderplugin loading?
17:12:38Bagderit loads, and runs. what more can it be? ;-)
17:13:22t0masBagder woke up this morning in a humor-mode? :P
17:13:37t0mashe wants to load a plugin from another plugin afaik
17:13:57Bagderthen he needs to exit the plugin and send a message in a queue that is read by the os
17:14:06Bagderto run another plugin afterwards
17:14:22Bagderor similar
17:14:43 Quit ac_learning ("using sirc version 2.211+KSIRC/1.3.12")
17:15:35Bagderfood time!
17:15:39t0masBagder: and there is sill not a simple way to launch viewer and after that delete the file right?
17:20:58HClBagder: i was more thinking of an exec() like function for plugins
17:21:25preglowsincerely doubt that's possible without hacking
17:21:26t0masmake that system() like for me...
17:21:34t0masas I want it to return :)
17:21:53amiconnt0mas: Returning is impossible.
17:22:06t0masok, recall me after it with a param then...
17:22:15amiconnThere is only one memory region for loading plugins, and they're linked to that fixed address
17:22:16HClpreglow: it should be easy..
17:22:23t0masor call some function pointer I give?
17:22:50preglowHCl: read what amiconn says
17:22:57amiconnChaining another plugin would _replace_ the first plugin, so there's nothing to return to
17:22:58HClpreglow: i'm well aware of that.
17:22:58preglowHCl: then repeat 'it should be easy', please ;)
17:23:13HClit should be easy :)
17:23:14amiconnChaining should be simple
17:23:22preglowHCl: you would have to make the plugin format relocatable
17:23:32HClpreglow: nope
17:23:37preglowHCl: if you'd like to work on that, i'll pat you on your back for sure, that would rock
17:23:45amiconnJust add a function to the core that can be called by a plugin to tell it "run this plugin after I return", then simply return
17:23:46HCli said exec() not system.
17:23:54t0masamiconn: and chaining two plugins? so this way: A(file) -> B(file) -> A(done)
17:23:57HClexec replaces the program currently running.
17:23:58preglowthat's possible
17:24:13preglowthen just do what bagder says
17:24:19preglowthat sounds like an exec() to me
17:24:40 Join |it|ChasKi` [0] (
17:24:50HCli rather have some more control over it and have the actual plugin do an exec() call rather than having a queue and the os handle it..
17:25:07preglowwhat's the difference?
17:25:28HCli don't think we even need a queue for this o.o..
17:25:42HClyou just load the parameter for the plugin in memory, probably iram
17:25:54 Join DMJC [0] (
17:25:57preglowrockbox already has the queue in question, no?
17:26:06t0masthat was my question
17:26:07amiconnWell, you can't load the new plugin from within the old plugin's code, cause you would overwrite the running code..
17:26:25HCldoes it?
17:26:29preglowof course
17:26:32t0masofcourse... you'll overwrite the JMP after the loading... so that's never done...
17:26:33preglowit might, at least
17:26:40amiconnSo you need a core function to do it
17:26:42HClamiconn, you'd have to load a routine in iram to do it, or simply have the function that actually does the exec in the firmware
17:26:44preglowdepends on size, i guess
17:26:54HCli meant does it have the queue.
17:26:55amiconnIram would be a real waste here
17:27:04t0masHCl: 1 second
17:27:13HClt0mas: lag between laptop and server
17:27:17HClnot actual irc lag
17:27:25t0masbut if that queue is already there...
17:27:33HClif the queue's already there thats fine
17:27:57HClbut otherwise
17:28:18HCli'd prefer storing the parameter somewhere temporarily, and using a routine to overwrite and call another plugin with that parameter
17:28:29t0masjust hilight Bagder like now and wait till he has finished eating so he can tell us :)
17:29:08amiconnHCl: Yes, and that routine must be part of the core, if we don't want to waste iram for that.
17:29:15HClamiconn: yea
17:29:21HClit shouldn't be too big
17:29:54amiconnWhether it's implemented like my 1st idea (tell the core what to call afterwards, then return) or like exec(9 shouldn't matter
17:30:03HClyea, i guess.
17:30:18DMJCamd 64 dualcore benchmarks are out
17:30:19HCli want to implement databox using that, so we can keep the searchengine seperate
17:30:38HClin order to enable in menu default searches like what the rio dj does
17:30:47HClDMJC: mm?
17:31:13t0masif you find out HCl... hilight me...
17:31:30t0masI still want to abuse the viewer plugin for displaying dictionary data
17:31:40t0maswould save me a lot of formatting work :)
17:31:48HClwhy does it have to delete the file afterwards anyways?
17:31:52DMJChas the dualcore desktop chip review up
17:31:56amiconnHCl: Just pu another special sub-browser in the menu somewhere, showing those search files, like the already existing browsers for plugins, fonts etc
17:32:04HClamiconn: yea.
17:32:05t0masHCl: it can be in /.rockbox/dict/
17:32:07t0masor something
17:32:19DMJCUp next are some gaming tests, which will essentially serve to illustrate the futility of running a dual-core processor in a single-threaded application.
17:32:30DMJCpfft... should be running ut2k4 on linux..
17:32:40DMJCthat uses threading..
17:33:05amiconnt0mas: Instead of using the viewer plugin for that, I might have a better suggestion
17:33:14t0masgo ahead? :)
17:33:31amiconn(chaining plugins will definitely need additional error checking, plus it causes an extra spinup etc)
17:34:02amiconnPut the formatting code into the plugin library, and use it in both plugins
17:34:16t0masthat would be a great idea...
17:34:26t0masand include the patch for the viewer in the same run? :)
17:34:28amiconnThat's what the plugin lib is for - reusable code
17:34:42HClThe X2 line will range from expensive to painfully expensive to root-canal-without-anesthetic expensive.
17:34:45HCl :P
17:35:36DMJCI wonder how those chips perform on linux
17:36:00HCli wonder wth people need that much speed fo?
17:36:02HCl :p
17:36:11DMJCgentoo heh]
17:36:12HClmy xp 1500 is still fine, really.
17:36:27DMJCand enlightenment
17:36:31 Nick StrathAFK is now known as Strath (
17:37:03DMJCnah, basically you use it for 3d and gaming
17:37:26HCli used to be into 3d till i realized its much easier to draw normally then shade with the computer
17:38:33DMJCI use 3d because I enjoy editing game content
17:38:40DMJCI probably wouldn't use it for art
17:41:58 Quit webguest26 ("CGI:IRC")
18:01:35preglowmy 2001fp is in copenhagen right now, and dell wants me to believe they have it here within the day
18:02:20*rasher goes to copenhagen and steals it
18:02:43*preglow too goes to copenhagen to kill the perpetrator
18:07:10*t0mas stole rasher's car with the screen in it :P
18:07:16t0masbad luck preglow :P
18:07:44rasherNow that'd be quite a feat, as I don't have a car
18:07:54t0masshit :P
18:08:02bobTHCand t0mas not have licence !
18:08:08t0massssst ;)
18:08:39*t0mas drove a lot of framing things around a few years ago...
18:08:50t0masworked at a flower company
18:09:04t0masframing -> farmer thins
18:10:09bobTHCwhich kind of flower ?? ;)
18:10:39t0masno, not weed...
18:11:11bobTHCtulip for sure ;)
18:11:23t0masand small trees
18:11:44t0masbut it was really bad work...
18:12:04t0masgetting there at 8:00 am on saturday :X
18:12:08t0mashalf alseep...
18:12:15t0masdamn... typing bad todat
18:19:38 Quit |it|ChasKi` (Read error: 110 (Connection timed out))
18:30:37***Saving seen data "./dancer.seen"
18:55:17 Join _aLF [0] (
18:58:30bobTHChi french ET !
19:04:57HClblerf @ the selfishness of some people.
19:05:32HClmeh, just some people
19:05:33HCl 1. Caring supremely or unduly for one's self; regarding one's
19:05:33HCl own comfort, advantage, etc., in disregard, or at the
19:05:36HCl expense, of those of others.
19:05:41HClwrong paste
19:05:51HCli meant that one
19:05:57HClgo there :/
19:15:14bobTHCit's not really new, africa is a place where slavery is always a big reality, it's will be the same when u talk about diamond, gold, oil, export farming .......
19:16:27bobTHCbut of course it's a shame for all of us who dont care about that
19:17:40preglowi care, it's just that it's more of a full time job keeping track of the status of every product out there, unless someone does it for you
19:17:42*HCl is searching for a chocolate milk brand thats slavery free
19:17:52HClpreglow: its been done :P
19:17:57preglowfor chocolate, yes
19:18:08HClyou can't do more than what you can do.
19:18:08preglowi almost never eat the stuff anyway
19:18:23HCli had chocolate milk occassionally
19:18:27HClbut i need to switch to a safe brand
19:18:37HClso i'm not having any till i find one
19:18:54preglowi seriously doubt they list norwegian brands anyway
19:18:55bobTHCchoose a "max haveelar" brand
19:19:01HClbobTHC: yea..
19:19:16HCldroste is a big one here, at least
19:19:25rasheristr something about this max havelar thing being a big scam, not sure what it was though
19:21:17bobTHCfor some it's a big scam because they use a part of the incomes to do commercials , but without commercials u cannot sell in our countries
19:21:54bobTHCand without selling u dont redistribute any part of the income to farmers
19:25:12*HCl stares at the last cvs commit
19:25:20HClnothing serious is happening, yup
19:25:35*HCl goes to grab his cat
19:25:48HCloh wait. nm o.o
19:25:55*t0mas doesn't want to no what you're going to de
19:26:08bobTHCwhen north ppl will be educated to "buy fair" they'll dont have to do commercials
19:27:17bobTHCbut it's not for tomorrow
19:28:24amiconnHCl: Didn't know there's something to stare at... ;)
19:28:26 Join Stryke` [0] (~Chairman8@resnet-241-86.resnet.UMBC.EDU)
19:28:37HClamiconn: its a bit of an idle commit :p
19:28:54HClt0mas: xD pet him. but he's sleeping under my blanket
19:29:35 Quit bobTHC ("Smoke Weed Every Day !")
19:42:12*HCl ish bored
19:43:06 Join |it|ChasKi` [0] (
19:43:36preglowfix dumb!"
19:45:12HCli was more thinking of taking a nap o.o
19:45:15HCland reading my book o.o
19:49:14preglowsomeone big band's playing the star wars theme outside
19:50:30preglowthe imperial march, to be more specific
19:50:38preglownot what they usually play around these days
19:53:16 Join webguest35 [0] (
20:08:18 Join edx [0] (
20:09:26 Quit Nibbler ("blubber")
20:12:22 Nick Gibbed is now known as Rick (
20:16:48 Quit |it|ChasKi` (Read error: 145 (Connection timed out))
20:30:38***Saving seen data "./dancer.seen"
20:36:57t0mas0x80484c0 in a hex editor is 80 48 4c 00 right?
20:37:04t0masthe extra zero can be added?
20:37:10t0masor am I doing something stupid then?
20:38:32preglowzero is added on the top
20:39:00preglowand the way it looks in a hex editor depends on what endianness the architecture uses
20:39:50t0maslittle endian
20:39:59t0mas08 04 84 c0
20:40:02t0maszo like that?
20:40:18t0mas(a 4 byte return address)
20:41:48preglowother way around
20:43:49 Join |it|ChasKi` [0] (
20:45:45t0masthe first one was right?
20:46:38preglow00 4c 48 80
20:46:49preglowc0 84 04 08
20:46:50preglowi mean
20:46:59preglowfuck this
20:47:06preglowi'm reading the wrong sequence of numbers all the time
20:47:13t0maseh? reverse it?
20:47:15t0masah ok...
20:47:19t0masit came from gdb...
20:47:44preglowthe smaller part of the number comes first
20:48:08preglowso the least significant byte is first
20:48:10t0mas"90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 c0 84 04 08" <−− you have three guesses what I'm doing...
20:49:09preglowwhy doesn't rockbox switch off automatically anymore?
20:49:13 Join Tang_ [0] (
20:50:18rasherbecause of the settings reset recently?
20:51:07amiconnIt should still switch off. The default poweroff timeout is 10 minutes
20:51:29rasherYes, I just noticed
20:51:37rashermaybe preglow was expecting it sooner
20:55:24preglowi had it on for, oh, seven hours while i slept the other day
20:55:30preglowdidn't expect to find it still on when i woke, heh
21:12:01 Join XShocK [0] (
21:13:21rasherevening XShocK
21:21:11HClwhats new..
21:31:50 Join Sucka [0] (
21:32:08 Quit webguest35 ("CGI:IRC")
21:37:16 Quit Harpy (Read error: 145 (Connection timed out))
21:46:46 Quit Sucka ("a bird in the bush is worth two in your house")
21:51:01 Join kramerica [0] (
21:58:52 Quit edx ()
22:13:30 Part amiconn
22:13:48XShocKhi rasher
22:13:58XShocKI was fast. :)
22:14:33HCl :)
22:14:36HClsup xshock
22:14:54*HCl wants someone to get mp3 playback to work :/
22:15:22*HCl looks around for a knowledgable person.
22:15:31 Quit Tang_ ("Chatzilla 0.9.68a [Firefox 1.0.3/20050414]")
22:17:15 Join amiconn [0] (
22:18:36 Join TCK- [0] (
22:25:07 Join tvelocity [0] (
22:25:09*HCl finds his knowledgable person.
22:25:15HClbut now i forgot my question, darnit
22:25:48HClamiconn: i assume we have code somewhere to shuffle a playlist? where?
22:27:08 Quit TCK (Read error: 60 (Operation timed out))
22:28:04XShocKpreglow: you once said that you have some kind of audio framework?
22:28:11preglowXShocK: i did?
22:28:27preglowwhat context?
22:28:29XShocKi think it was you. :) something like half a month ago. :)
22:28:29amiconnHCl: There is code somewhere for that... but I don't know where. Should be somewhere in apps/
22:28:34HClokay :/
22:28:40XShocKthat you have something working..:)
22:28:45HClwanting to add it to the searchengine :)
22:28:45preglowi don't ;)
22:28:50HCland then i'll add the time limit thing..
22:28:59preglowlinus is the one who's got playback going
22:29:12HClwe haven't heard anything about that since, have we? :/
22:29:19preglownah, he's busy
22:29:33XShocKok. i might have forgotten something
22:29:34preglowbut it's not like anyone else is attempting it anyway
22:30:40***Saving seen data "./dancer.seen"
22:31:57 Join Sucka [0] (
22:36:19HCli really want it working so i can start on the runtime db though :/
22:36:47preglowyou can start on that anywya?
22:36:58HCli don't have easy access to a simulator
22:37:06preglowwhy not?=
22:37:10HCland i need to actually be able to access the wps and back
22:37:15HClin order to test properly
22:37:20HClbecause i code remotely on my server
22:37:25preglowso do i
22:37:28HCland send files by ftp back to my laptop
22:37:33HCland i don't like vnc :/
22:37:41HCli just want to develop on my iriver, heh.
22:37:52HClyea, but no simulator, obviously.
22:38:03preglowi use ssh and can use a sim with no problem
22:38:11HClremote x?
22:38:20HCli don't use that.
22:38:26preglownor do i, but i can for the sim
22:38:32preglowit's no problem, and completely transparent
22:38:37preglowi seldom use the sim anyway
22:38:38HCli'm not gonna set up an xserver on my windows comp just for the sim
22:38:42preglowsince that doesn't understand asm, heh
22:39:05Bagdercygwin has a fine xserver
22:39:07preglowwell, it's gonna simplify your coding, and it's that not much setting up, you just install cygwin, and there you are
22:39:11preglowbut i'm not forcing you
22:39:28Bagderpersonally I wouldn't dream of doing all that coding without using the sim
22:39:30preglowi sure wish i could use the sim
22:39:43*HCl ish happy with developing on his iriver.
22:41:10Bagdergdb is harder on target
22:41:21preglowand oh, i wish i had gdb sometimes
22:41:33HCli rarely need gdb o.o
22:43:20t0masBagder: that cygwin X server... I have it running... but how do I connect something to it? :)
22:43:41Bagderxeyes &
22:44:00t0masI ment remote ;)
22:44:09t0masjust define DISPLAY=myhost:0 ?
22:44:11HClset DISPLAY=<ip>:0
22:44:12Bagderah, run 'ssh -X host'
22:44:12t0masand it should work? :)
22:44:17Bagderand it'll do it by itself
22:44:25amiconnIt should, yes
22:44:25HCl0 being the display number
22:44:38t0masok, why doesn't it then? :P
22:44:39BagderHCl: that's not needed if ssh is used
22:44:58HClyea, well, i haven't figured out putty x session stuff
22:46:02preglowyou click enable x11 forwarding?
22:46:23amiconnt0mas: I once tried to connect from cygwin x to debian x and vice versa. For some reason it only worked in one direction, don't remember which one atm
22:46:23*t0mas slaps his cygwin setup...
22:46:37t0masamiconn: I'm using a debian server too...
22:46:43t0masso that might be the problem
22:46:55t0masI'll test tomorrow with my laptop... thats not debian
22:47:04amiconnI don't use a server, I just have a debian VM for occasional testing on linux
22:47:06preglowi sincerely doubt debian enters into it
22:47:30rasherDebian sets SSH up with X forwarding disabled
22:47:35rasherssh server, that is
22:47:41t0masnot mine
22:47:47t0masas I compiled it myself... wanted chroot options
22:47:49amiconnrasher: I didn't try to use ssh
22:48:07amiconn...just setting the DISPLAY
22:48:10t0masand I normally just remove the debian config files... and write/copy my own documented versions
22:49:02preglowamiconn: then you need to hack a config file in cygwin
22:49:37t0masBagder? "cpp0: config.h: No such file or directory"
22:49:51t0mas"gcc version 2.95.4" to old?
22:50:14amiconnstone age...
22:50:17t0mashow do I tell it to use /usr/local/gcc-3.3.4/ ?
22:50:41BagderI bet /usr/bin/gcc is a symlink
22:50:55t0mas"/usr/local/gcc-3.4.3/bin/gcc" that is
22:51:02Bagderadd that first in your PATH
22:51:03t0masyes, but debian is used to the old 2.95
22:51:23preglowdebian still uses 2.95?
22:51:26Bagdertogether with "/usr/local/sh-elf/bin:/usr/local/m68k/bin" ;-)
22:51:34Bagderpreglow: that's probably "stable" or something
22:51:40preglowoh, go figure why i never use stable
22:51:56t0masand ls on /usr/local/gcc-* gives a long list on mine...
22:52:00amiconnpreglow: my debian install uses 3.3.5
22:52:04preglow2.95 is pretty crap compared to recent 3.x.x
22:52:07t0masbut 2.95 is the system default
22:52:12Bagdergcc −−version
22:52:12Bagdergcc (GCC) 3.3.6 (Debian 1:3.3.6-3)
22:52:20preglowsarge, yes?
22:52:21amiconnt0mas: I just did a default install...
22:53:00t0masamiconn: gcc version 3.0.4 in woody
22:53:04t0masand 2.95
22:53:07t0masboth installed
22:53:22 Nick tvelocity is now known as tvelocity[away] (
22:53:29t0masand I added 3.3.x and 3.4.3
22:56:07amiconnI didn't use 'woody' (afaik. I always mix up these odd names. Why not just use version numbers?)
22:56:26rasherthere are version numbers
22:56:35t0mas3.0 is woody atm?
22:56:37BagderI don't care about them either, I use unstable
22:56:48*t0mas has stable for his server and router...
22:56:51Bagderit floats
22:57:41HCli had a mix of unstable with ubuntu on my laptop..
22:57:48rasherI moved my server from stable to testing once it "neared" release a couble of months ago
22:58:01t0masmy workstation is testing...
22:58:06t0masas I needed some new things...
22:58:12amiconnI checked what I have... I used the debian sarge net installer
22:58:15t0masbut I'm using winXP 90% of the time :)
22:58:18 Join arfo [0] (
22:58:21HClmy server is woody with sarge
22:58:39t0masand my laptop is just bad... fedora...
23:01:32 Part arfo
23:14:04amiconnHmm. Now both directions do not work for the x connection :(
23:14:46*t0mas will just boot to linux tomorrow morning...
23:14:59t0masjust have to read some articles... so openoffice can to that..
23:15:47t0masgood night
23:19:33 Join matsl [0] (
23:24:14 Quit Sucka ("a bird in the bush is worth two in your house")
23:26:19 Join Sucka [0] (
23:26:50 Join TCK [0] (
23:37:25 Join ashridah [0] (
23:44:47 Quit TCK- (Read error: 110 (Connection timed out))

Previous day | Next day