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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2005-10-06

00:22:54 Quit ashridah ("Leaving")
00:25:54 Nick TiMiD[away] is now known as TiMiD (n=TiMiD[FD@asgard.valombre.net)
00:25:54DBUGEnqueued KICK TiMiD
00:25:59TiMiDhi
00:26:25MoosBonsoir :)
00:26:36TiMiD'soir
00:26:50TiMiD(la facon snob ^^)
00:26:57Moos:D
00:27:07TiMiDt'es francais ?
00:27:18Moosoui, parisien ;)
00:27:23TiMiDwaaaa kewl :p
00:27:39Moosand you?
00:27:45TiMiDcomme ca jme sentirai moins seul quand je ferai des phrases qui veulent rien dire
00:27:51Moos:D
00:27:59TiMiDmoi je suis argenteuillais expatrie a compiegne
00:28:00Moosbienvenu au club :-)
00:28:07Moosohh
00:28:15Moospour le taf?
00:28:16TiMiDje fait mes etudes la bas
00:28:21Moosok
00:28:26TiMiD(derniere année en ingé)
00:28:44Moosbon courage, bientôt la fin ;-)
00:28:55TiMiDen fait plus que 6 mois a tirer
00:29:01TiMiDapres c'est le stage
00:29:13Moosta déja un plan pour ton stage?
00:29:14TiMiD(repos :) )
00:29:20Mooshehe
00:29:31TiMiDben je vais pe partir au japon chez cogema
00:29:44Moosbon plan
00:29:53Moosta des connexion la bas?
00:30:00TiMiD(ca rentre tt a fait dans le cadre de mes plans de domination du monde ca le traitement des dechets radioactifs ^^)
00:30:06TiMiDeuh
00:30:13TiMiDque dalle ^^
00:30:21Moos:-)
00:30:26TiMiDmais en fait y a un prof la bas qui s'en occupe
00:30:32Moosjaponais d'origine?
00:30:36TiMiD(lui il connait bien le japon)
00:30:38Moosa k
00:30:55TiMiDnan moi shui francais pure souche :p
00:31:05Moos:D
00:31:24Moosun gaulois :)
00:31:35TiMiDun romain ^^
00:31:40Mooshehe
00:32:01TiMiDc bizarre kon se soit pas fait kicker
00:32:11TiMiD(ils doivent dormir :D)
00:32:29TiMiDj'en ai vu se faire jarreter pourpas parler anglais :)
00:32:45TiMiDmais ca devait etre pr deconner parce qu'ils ont l'air assez cool ici
00:32:50Moosc pas la politique de la maison, mais vu qu'il ya personne ca ne les derrange pas, bien que nous poluons un peu les logs avec notre français :)
00:33:21TiMiDtu fait partie de leur equipe ?
00:33:52Moosnon je n'ai pas ce privilège (surtout pas les compétences) :D
00:34:00TiMiDok :)
00:34:39TiMiD(les competences ca dépend qui hein quand tu vois certains bouts de code tu te demande si c'est la biere ou l'exta... enfin bref :p )
00:34:45Moospar contre j'ai vu un peu l'évolution de ton taf à propos de la télécommande, ça à l'air prometeur
00:34:51TiMiDah
00:34:53Moos:D
00:34:57TiMiDmoi je suis pas firefly
00:35:12TiMiDc pas moi qui ai pose sur le forum
00:35:17TiMiDposte
00:35:22Moos?? tu ne tafé pas dessus aussi mais d'une autre manière?
00:35:27TiMiDsisi
00:35:43Mooset comment ça avance?
00:35:47TiMiDmoi je refait ca en faisant le decrassage de profondeur (tt recoder :p)
00:35:56TiMiDbonne question :)
00:36:03Moosok, va te falloir du courage
00:36:16TiMiDen fait la j'ai l'affichage des fichiers qui marche en mode normal et id3db
00:36:19TiMiDsur tt les modeles
00:36:36Mooshé! c'est un trés bon début
00:36:43TiMiD(c deja pas mal vu que j'ai refait 90% du code d'affichage "from scratch")
00:36:55Moosc clair
00:37:23TiMiDapres c pas tres utilisable vu qu'il manque le wps et que quand tu quitte les menus, le remote reste comme si tu y etais encore
00:37:37Moosok
00:37:45Moospour l'instant :)
00:37:47TiMiDune fois que j'aurai bien teste tt ca je vais essayer de soumettre un patch
00:37:56Moosce serait cool
00:38:08TiMiDsi il est accepte oui ce serai pas mal :p
00:38:10Moosje suis sur que cela pourrait itéressé Linus
00:38:18Moos*intéressé
00:38:29TiMiDmm si c juste les fichiers c deja un debut et ca derange personne
00:38:44TiMiDLinus a l'air occupe en ce moment
00:38:45Moosc clair
00:38:52Moosoui on dirait
00:39:05TiMiDdc va falloir soudoyer kk1 d'autre :p
00:39:14Moosplusieurs patch sont toujours en attente de son assentiment
00:39:17Moos:D
00:39:21TiMiDoulalal :(
00:39:56Moosmais les patch sont fait, il faut juste su'il y jette un coup d'oeil final
00:39:57CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:39:57*TiMiD espere que son patch pssera avant celui du dircache parce que ce patch la casse tt
00:40:04Moosavec ses yeux persans :D
00:40:23TiMiD:)
00:40:30Moosle patch de Slasheri est prés à être soumit
00:40:44Moosles patch de Xavier aussi
00:40:52Moospour le browser de jpeg
00:40:56Mooset la radio
00:41:08TiMiDtt que ca touche pas a ce quej'ai modifie ... :D
00:41:18Moosheureusement, oui
00:41:20Moos:)
00:42:02TiMiDben celui de Slasheri modifie un peu des trucs que j'ai modifie mais c pas grave c'est pas non plus 100000 lignes de code qui changent, je devrais pouvoir gerer ca a la main
00:42:17Moosca va :)
00:42:37TiMiDouais ca devrais passer si je me bouge un peu le Q
00:42:55Mooson compte sur toi ;-)
00:43:31TiMiD(la en ce moment j'ai la motivation qui retombe un peu et je passe mon temps au bar etudiant a boire des pintes avec mes potes alors rb... ca avance - vite forcement :D)
00:43:41TiMiDoops g brise tt les espoirs :)
00:43:46Moos:D
00:43:54TiMiDnan mais je vais me le finir ce truc
00:43:56TiMiDau - ca
00:44:04TiMiDque j'ai au - un truc a mettre sur mon CV
00:44:13Mooshehe :)
00:44:32TiMiDautre que le tout plat "a fait ses stages et ses diplomes"
00:44:45Moosc clair
00:44:52Moosc original
00:45:02TiMiDc meme terrible :)
00:45:08Moos:D
00:45:10TiMiDca montre a l'employeur que tu aimes ca ^^
00:45:42TiMiDca peut largement te faire passer devant un autre
00:45:44Moosoui, de plus ça te sert personnelement ;-) avec ton iriver
00:45:55TiMiDaccessoirement ;)
00:46:09Moosoui, ca peu etre un détail qui peut aider
00:46:19TiMiD(mais le patch de firefly marche pas mal non plus, meme si c'est un peu crade niveau code)
00:46:49Moosoui, mais ça ne peu pas être une version finale
00:46:57TiMiD:/
00:47:02Moosc'est trés bien en attendant
00:47:09TiMiDje me demande pourquoi il continue a l'updater dailleurs
00:47:21TiMiDmoi si il veux coder j'ai du boulot pour lui
00:47:23TiMiD:)
00:47:29Moos:D
00:47:53TiMiD(commencer par remplacer les 4000+ appels de fonctions d'affichages maintenant desuets :p)
00:47:56Moosil a le mérite d'avoir eu l'intention au moins, on ne peut pas lui reprocher ;)
00:48:13Mooshehe :)
00:48:15TiMiDah mais je lui reproche rien, je trouve ca meme tres bien au contraire
00:48:30TiMiDcomme ca il connait un peu mieux le code et il pourra m'aider :)
00:48:37Moosc clair, beaucoup d'users sont dépendant de la remote
00:48:47TiMiDben moi ou j'habite ...
00:48:49preglowplease, boys
00:48:51preglowenglish in here
00:49:08Moosoops hey preglow, scuse
00:49:12TiMiDsi j'ai pas le remote pour faire genre cle usb c mort, c pique en 2sec
00:49:15TiMiDoops
00:49:17TiMiDscuse :)
00:49:37Mooslot of users don't using Rb cause the remote
00:49:50Moospersonaly since Rb I don't use it
00:50:19Moosbut with your works, sure I will
00:50:47TiMiDsure :) and you will be very happy when it will crash right in your hands
00:50:54Moos:D
00:51:04TiMiD(yes, you may have noticed, I'm very optimist :p)
00:51:16Moos:)
00:52:03Moospv for french ;-)
00:52:15TiMiD^^
00:58:33TiMiDpreglow: are you a dev ?
00:58:42TiMiD(I have a question :) )
00:59:00Moossure he is, a codec man :D
00:59:11TiMiDperfect :)
01:00
01:00:35TiMiDwell
01:00:48TiMiDit's about the header in the source files
01:00:58TiMiDwhn you put your name
01:01:09TiMiDhow do you handle that ?
01:01:41TiMiD(when the file is 100% from you, it's not a problem, but when yu picked up code from other files ....)
01:02:55preglowwell
01:03:04preglowthus far, i've just put my name there if the file is purely mine
01:03:18preglowthat is, if i've written in from scratch
01:05:31TiMiDin my case (remote support), there is some code I picked up in other files, because it did the same thing mainly the same way (but these files will be removed if my patch is accepted)
01:06:05TiMiDI don't think it's fair to say it's 100% my work
01:06:17preglowthus far, my name is only in four source file headers, it seems
01:06:31preglowwell, it's up to you
01:06:35TiMiDI will put their names too, but I never saw that in other files of rb
01:06:56TiMiDthat's why I ask :p
01:07:05preglowi don't really know much about his
01:07:11preglowi'd ask some of the core devs
01:07:16TiMiDok thx :)
01:08:26TiMiD(linusN is concerned, he is one of those I stealed code ^^)
01:08:42Moos:-)
01:11:07preglowcan't say i worry so much about this myself
01:11:29preglowyou can steal as much of my code as you please, and claim copyright yourself without me killing you
01:11:58preglowbut then i don't believe my code is very relevant to you, hehe
01:12:08 Join amiconn_ [0] (n=jens@p54BD6DB2.dip.t-dialin.net)
01:12:41TiMiDof course Iwill never put my hands in your assembly
01:12:56TiMiDyou don't have to worry about that :)
01:13:32TiMiD(and I'm pretty sure only a few devs are going to modify your codecs :) )
01:13:36TiMiDhi amiconn_
01:13:48TiMiDbtw, if someone wants to test
01:13:50TiMiDhttp://timidzone.free.fr/pub/rockbox/remote/rockbox.zip
01:13:54TiMiDhere it is
01:14:06TiMiD(build with my remote patch)
01:16:32***Saving seen data "./dancer.seen"
01:17:18Mooshehe :P
01:17:36Moosneed to sleep here
01:17:45Mooshave a good night guys
01:17:46TiMiDI will go too
01:17:52Moosc you later
01:18:02TiMiDgood night !
01:18:08Moosbonne nuit ;-)
01:18:09TiMiD(bone nuit si j'ose dire)
01:18:12Moos:D
01:18:20 Quit Moos ("Glory to Rockbox")
01:18:21TiMiDroh je vais me faire kicker maintenant
01:25:01preglowhaha
01:25:19preglowthere are already quite a few devs like me doing assembly work on codecs, so who knows
01:27:30TiMiDCodecs are the only part of rockbox I truly don't want to understand :)
01:28:09TiMiD(too much aths behind)
01:29:39 Quit amiconn (Read error: 110 (Connection timed out))
01:29:39 Nick amiconn_ is now known as amiconn (n=jens@p54BD6DB2.dip.t-dialin.net)
01:31:19preglowweell, might be true, depends on what you want to do
01:31:29preglowi haven't done any codec work involving maths thus far, but i am capable of doing so
01:31:39preglowhell, i didn't even WANT to do codec work when i joined, it just happened ;)
01:32:11 Join Paul_The_Nerd [0] (n=paulthen@cpe-66-68-93-2.austin.res.rr.com)
01:32:32TiMiDyou lost a bet? :D
01:47:15 Quit k-man ("Client exiting")
01:50:36TiMiDjust a question about remote : how do I handle pgup/pgdown since the screens size is different ? :p~
01:51:32TiMiD(this is a pbl :) )
01:52:05TiMiDif the jump is good for the remote, it will be too small for the main display
01:52:11TiMiD...
01:52:34Paul_The_NerdHow 'bout this:
01:52:52Paul_The_NerdIf they use the buttons on the box, use it's distance
01:53:03Paul_The_NerdIf they use the buttons on the remote, use the smaller one.
01:53:13Paul_The_NerdMost likely they're looking at the screen that aligns with the buttons they're pressing.
01:54:00TiMiDgood solution :)
01:54:09TiMiDthx !
01:54:28Paul_The_NerdNo problem. Seems like a reasonable enough way.
01:54:41Paul_The_NerdAt least that way I'd like it. :)
01:55:44TiMiDI did it in a way that each screen gets desynchronized when pgup/pgdown (yes, you can call it a bug)
01:55:47TiMiD:)
01:57:18preglowTiMiD: nah, no bet, i just suddenly wanted to do some assembler
01:57:20preglowTiMiD: i'm mad
01:59:40 Quit ]RowaN[ ()
02:00
02:03:08 Join mike8901 [0] (n=m@ool-457804fa.dyn.optonline.net)
02:04:56Paul_The_NerdBut clearly mad in that *good* sort of way, preglow.
02:07:10TiMiDholy/mad assembly optimizer :)
02:07:16Paul_The_NerdHrm
02:09:46Paul_The_NerdIs it going to be possible to have the two screens completely independent eventually?
02:10:31Paul_The_NerdThough I can see all kinds of entertaining problems coming up from that sort of thing anyway.
02:11:57preglowhaha
02:12:13preglowit would rock, though
02:12:23Paul_The_NerdAbsolutely
02:12:53Paul_The_NerdUntil people got the clever idea of trying as hard as they can to select two different files (possible with differently associated viewers, or just a viewer and a music file) as close to simultaneously as they could.
02:13:01TiMiDI implemented it in that spirit
02:13:39Paul_The_NerdThough I would like to see it kick the remote to WPS whenever I run a plugin, just because I have dreams of being able to control music while playing various games.
02:13:56TiMiDeven if multithreading the main app is not possible currently, the "widgets" I did were coded to work like this
02:14:02Paul_The_NerdGotcha
02:15:07TiMiDeach list in both screens is independant (there is some wrapper fn to synchronize them)
02:15:55Paul_The_NerdSo, you said they get desyncrhonized currently. How do you handle that?
02:16:35TiMiDthey are not desynchronized anymore :)
02:17:25Paul_The_NerdAh
02:17:27Paul_The_NerdOkay
02:18:24TiMiDwell if you want to know how it works, the main program uses a synchronized api that uses the api of unsynchrinized lists and make them synchronized
02:19:22TiMiDif you want unsynchronized lists, you just have to call the desynchronized api instead
02:20:16 Quit Bagder (Read error: 110 (Connection timed out))
02:20:35Paul_The_NerdSo, it's ready to be desyncronized should someone want to.
02:20:58TiMiDyes
02:21:32TiMiD(but in the present state of rb code, this would involve rewriting a lot of things)
02:21:45 Join bagawk [0] (n=lee@unaffiliated/bagawk)
02:21:49Paul_The_NerdFair enough
02:26:11TiMiDoh also I noticed that the putc fn wasn't implemented on remote display (not a big deal since nobody uses it ;) )
02:30:31 Join XavierGr [0] (n=XavierGr@ppp15-adsl-30.ath.forthnet.gr)
02:30:39TiMiDhi XavierGr
02:33:11XavierGroh hi TiMiD!
02:33:33XavierGrI read the news that you released a build with your remote?
02:33:43XavierGrDoes it have any bugs?
02:34:54TiMiDhmm
02:35:01TiMiDI've better to reupload it :)
02:35:14TiMiDcurrently it's running on my iriver for 3 days
02:35:15XavierGrwhy?
02:35:17TiMiDno crash
02:35:19XavierGrbugs?
02:35:27Paul_The_NerdWell, on my box it summoned forth the angelic hosts and the demons of hell, who then proceeded to do battle for the immortal soul of my h120
02:35:28XavierGrwell that it!
02:35:29TiMiD(but I fixed a small bug right here)
02:35:31XavierGr^that's
02:35:36Paul_The_NerdIt was fun to watch though, so I'm not sure if it's a bug, or undocumented feature.
02:36:09XavierGrI just made a biuld with jpeg file scrolling, fm preset, dircache and firefly's remote patch!!
02:36:31TiMiDgood :)
02:36:33*XavierGr *sighs* if only a dev was willing to commit my work...
02:36:59TiMiDPaul_The_Nerd: bug ?
02:37:11Paul_The_NerdIt was a joke, TiMiD.
02:37:34TiMiDok :)
02:37:45Paul_The_NerdI did try your build briefly, didn't encounter any problems, but went back to the one using Firefly's patch for the time being.
02:37:54TiMiD(since I'm not native speaker I have some understandings pbl :D )
02:38:05Paul_The_NerdAh, I understand.
02:38:14TiMiDyes, that's the best thing to do
02:38:24Paul_The_NerdXavierGR: Is this off of the latest CVS too?
02:38:25XavierGrTiMiD are you from Frace?
02:38:28XavierGr^France
02:38:31TiMiDoui :)
02:38:43TiMiDnihaha
02:38:51XavierGrSo that's why you flooded the logs with French.....
02:38:53XavierGr:P
02:39:31Paul_The_NerdYeah, I started reading the log, and for a few pages I just had no idea what was going on.
02:39:32XavierGr2 commitments before I think. Without the sudoku commits.
02:39:47TiMiDouch :)
02:40:25TiMiDit's true that I'm more fluent when I don't need a dictionnary :p
02:42:59TiMiDwell I've got another question
02:43:51TiMiDdo you think it's better if pgup/down returns to the beginning when reachin the end or if it stays on the last item ?
02:44:14Paul_The_NerdI think that if it is not on the last item, it should go to the last item.
02:44:20TiMiDok
02:44:22Paul_The_NerdIf it is on the last item, it should go to the first.
02:44:28Paul_The_NerdFor PgDn
02:44:38TiMiDsame for pgup :)
02:44:41Paul_The_NerdBut that's what I'd expect it to do. I'm not sure if others would agree.
02:45:37TiMiDno, it's a good behaviour (I'm not very used to gui design ^^)
02:49:12XavierGrso did you finished tree.c ?
02:49:47 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
02:50:22TiMiDyes
02:50:41TiMiDit also seems to work with id3db
02:51:06XavierGrand now you are working on?
02:51:11TiMiD(well there is one thing missing : button bar)
02:51:40TiMiDI'm doing bugfixes (when I have the time)
02:52:35Paul_The_NerdButton bar?
02:52:40TiMiDwhen buttonbar will be done (used on archos recorders) I will submit a patch
02:53:18TiMiDyes, the contextual bar above F1 F2 F3 on archos
02:53:32Paul_The_NerdAh
02:54:08Paul_The_NerdI'm not terribly familiar with them. My younger brother has one that he never uses. I'm trying to convince him to give it to me, or at the very least loan it indefinitely or sell it cheaply.
02:54:17XavierGrbut the recorders haven't got a remote...
02:54:37XavierGrarchos models I mean
02:54:57TiMiDît must be handled anyway
02:54:58Paul_The_NerdXavierGR: It's a rewrite of the system for an arbitrary amount of displays, so the way it handles the primary display is rewritten too, if I understand correctly.
02:55:29TiMiDPaul_The_Nerd: I couldn't have explained it better :)
02:56:03XavierGrit must great job I wish for a quick commit.
02:56:09XavierGr^be
02:56:46XavierGrwhat files have you changed so far?
02:57:05TiMiDsure it's a great job if it's me who is doing it hmmhmm ;)
02:57:27TiMiDhttp://timidzone.free.fr/pub/rockbox/remote/
02:57:48TiMiDhere are all files rewritten / modified / newly created
03:00
03:02:41XavierGrwitch is the syntax to undo a patch?
03:02:59TiMiDdon't know :(
03:03:53XavierGrwell I will search it on rockbox, I am sure it is there.
03:09:02XavierGrpatch -p1 -R < patchfile
03:09:11XavierGrthat's it!
03:09:20TiMiDgood thing to know :)
03:11:08XavierGrwell it is getting late here!
03:11:13XavierGr4.11 am
03:13:21TiMiDwhere ? grece ?
03:13:34XavierGryup!
03:13:56TiMiD=(^o^)=
03:14:02 Quit preglow ("leaving")
03:14:10TiMiD(not very hard with the Gr :p)
03:14:29XavierGrLol I just use it because Xavier is already taken.
03:14:46 Nick TiMiD is now known as TiMiDfr (n=TiMiD[FD@asgard.valombre.net)
03:14:46DBUGEnqueued KICK TiMiDfr
03:14:53TiMiDfrthen you are not alone :)
03:15:49TiMiDfrI learnt antic greek at school, but I can't remember a word ^^
03:16:12TiMiDfrmaybe only some parts of the alphabet
03:16:33***Saving seen data "./dancer.seen"
03:17:45XavierGrWell i cant say that I understand ancient Greek but there are major similarities in ancient and modern Greek.
03:18:12XavierGrthe alphabet is the same, many words, but the detail is lost in modern Greek.
03:18:29TiMiDfryou learnt it ?
03:19:55XavierGrWell they are trying to learn this at school but anyone that has a practical mind, and follows sciences of this subject (not philosophy, philology e.t.c) will soon forget them.
03:20:24XavierGrYou can understand some things partly even if you've never learnt it before.
03:20:31 Join slaggathor [0] (n=47f7e6b4@labb.contactor.se)
03:21:40XavierGrAncient Greek is a very detailed language, it has rules for everything, to the tiniest little detail.
03:21:53slaggathorTiMiD +Xavier −− the remote work is great, that's really the only thing keeping me from using rb all the time ^_^
03:22:00TiMiDfrit's somwhere the same thing when you are french since a lot of greek words are roots of modern french words
03:22:06 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
03:22:21XavierGrwhy me slaggathor?
03:22:22TiMiDfrbut grammwar was ... arghhhhhhhhhhhh
03:22:39XavierGrI didn't do anything (except the old remote patch)
03:22:42TiMiDfrXavierGr: you also did a patch once ago
03:22:46slaggathorXavier −− yeap
03:22:59XavierGrTiMiDfr: I agree grammar was ...
03:23:19XavierGrI remember that they tried to learn us some french in school too...
03:23:26TiMiDfrouch :)
03:23:28XavierGrwell they failed! :P
03:23:38TiMiDfrfrench sucks
03:23:48XavierGrI can't remember a think except one or 2 phrases.
03:23:53TiMiDfr(and I'm well placed to say that :) )
03:23:56slaggathorI like how pretentious people try to make pseudo-latin plurals out of words that are actually greek
03:24:17XavierGrwell you can't actually say suck.
03:24:32XavierGrslaggathor: like?
03:24:36TiMiDfrall european languages are too much complicated
03:24:48TiMiDfrwhen you compare with chinese / japanese
03:24:54slaggathornot rhinoceri but rhinocerontes... not octopi but octopodes
03:25:12XavierGroh they say such things...
03:25:24TiMiDfrthe grammar is ridiculously easy in those asian languages
03:25:28slaggathorya chinese and japanese are easy (except for the politeness)
03:25:51XavierGrI thought they were complicated languages and difficult to lear..
03:26:20TiMiDfrexcept when you are used to complicated sentences forms like the one we use to speak
03:27:02slaggathorgrammatically they are quite simple, japanese is like german always sticks with the rules. but the pitches/tones in chinese are a little hard to learn
03:27:08TiMiDfr(politeness is also a little difficult to understand, even for the japanses themselves, so always use the polite form ^^)
03:27:16slaggathor^_^
03:27:48 Join Bagder [0] (n=daniel@1-1-5-26a.hud.sth.bostream.se)
03:27:51TiMiDfrthere are far less grammar rules in japanese than in german though
03:27:55XavierGroh TiMiDfr strange that you say that French sucks. Usually French people are known for their stubborness to speak only French even they know another language.
03:28:08XavierGrOr so I've heard can you confirm this?
03:28:20TiMiDfrmuhahaha mais si je parlais francais vous comprendiez rien bandes de moules :)
03:28:36XavierGrok I will use babel fish
03:28:46TiMiDfrha I will try too
03:28:54slaggathornothing can be worse than the catalan ppl in barcelona refusing to speak spanish...
03:28:57TiMiDfrshould be funny ;)
03:29:55TiMiDfrs/comprendiez/comprendriez
03:30:16TiMiDfrtaht gives us "muhahaha but if I spoke French you would include/understand anything bands moulds :)"
03:30:19XavierGrwell yes babel didn;t translated that..
03:30:43slaggathorheh
03:31:08XavierGrwhats bands moulds?
03:31:22TiMiDfrhmm
03:31:31TiMiDfrI don't know how to say it in english
03:31:37TiMiDfrhow to explain ...
03:31:41XavierGrlol it is funny to say
03:32:07TiMiDfrit's like "guys"
03:32:57XavierGrso you said: if I spoke French you wouldn't include/understand anything guys??
03:33:28TiMiDfrbut with a "you are not clever" connotation (nothing not kind though :) )
03:33:46TiMiDfrsmthing like this :)
03:34:12slaggathorfunny that altavista translate as moulds, isnt it mussels? which is even funnier...
03:34:20TiMiDfr"bande de moules" is smth I say to my friends when they do silly things :)
03:34:27XavierGrJe ne sais rien au sujet de desole francais !
03:34:36TiMiDfraouch :)
03:34:42XavierGrdid that came right?
03:34:48TiMiDfr:/
03:35:04XavierGrwhere is the sorry remark?
03:35:10TiMiDfrI understood but ...
03:35:23XavierGrI don't know anything about French sorry!
03:35:27TiMiDfryou want to say "sorry I don't understand french" ?
03:35:30XavierGrthat;swhat I translated.
03:35:32TiMiDfrok
03:35:44XavierGrah I should have put a comma
03:35:48TiMiDfrJe suis esole, je ne parles pas frrancais
03:35:53TiMiDfrdesole
03:36:23TiMiDfr(automatic translation sucks a bit :p)
03:36:44XavierGra lot!
03:36:58XavierGrFrench language is difficult to pronounce.
03:37:14TiMiDfrenglish too :)
03:37:31slaggathormy favorite: j'ne comprends pas rien
03:37:37TiMiDfrjapanese is easy for french
03:37:44TiMiDfrouch :)
03:37:52TiMiDfr"Je comprends rien"
03:37:54XavierGrI can imagine.
03:38:03TiMiDfr"Je pige que dalle"
03:38:09slaggathorya I know my version was deliberate
03:38:12slaggathor^_^
03:38:33TiMiDfrso :)
03:39:03slaggathorit's funnier that way / I sound stupider that way
03:39:31TiMiDfryou made the sentence yourself ?
03:39:53XavierGrI measuring rod which flagstone −−-> lol what that supposed to mean?
03:40:05XavierGrhahaha!
03:40:20slaggathoryes! o_O
03:40:29TiMiDfrlol XavierGr
03:40:43XavierGris the transaltion right? I don't think so.
03:40:49TiMiDfrit meant the same thing that "Je comprends rien"
03:41:08XavierGrcompletely irrelevant!
03:41:08TiMiDfr>> "I dont understand"
03:41:13slaggathor"I don't understand nothing"
03:41:15TiMiDfryes :)
03:41:47XavierGryeah I now but while the first sentence was translated right the second shoed up this!
03:41:58TiMiDfrslaggathor: I assume you are right :)
03:42:13TiMiDfrXavierGr: the first is how you find it in books
03:42:18slaggathorno I am so wrong, that's the whole point of saying it like that ^_^
03:42:35TiMiDfrthe second how you would say when you speak to friends
03:42:56XavierGrwell in Greece we have some saying to make jokes for people that do not speak English very good.
03:43:00TiMiDfrslaggathor: you are native english speaker ?
03:43:10XavierGrE.X: I tell England very best! :X
03:43:10slaggathorouais
03:43:22TiMiDfrkewl :)
03:43:44TiMiDfrXavierGr: same in france :)
03:43:58XavierGr:D
03:44:06TiMiDfr"Je parler tres bien la france"
03:44:15XavierGrOr sometimes we translate Greek phrases to English ones.
03:44:38TiMiDfrlike ?
03:44:55XavierGrE.X: You will eat wood -> I will bit you up. (yes I know it is hilarious but we use it).
03:45:16slaggathorhee
03:45:31XavierGrbecause in the old days some mothers were punishing their kids with a thin wood.
03:46:42slaggathorin the not-so-old days, our mothers would say "your ass is grass, and I'm the lawnmower"
03:46:54XavierGrhaha!
03:47:18TiMiDfrho :)
03:47:49TiMiDfris there any sexual connotation ?
03:48:06slaggathorI dont think so...?!
03:48:15TiMiDfr(if I translate it in french it's .. ambiguous)
03:48:16XavierGrIf the discussion was between husband and wife maybe...
03:48:23slaggathorhehe
03:48:55TiMiDfrwell I won't make you a drawing :)
03:49:03slaggathorit just means "I will beat the hell out of you"....
03:49:20XavierGrnow that I am thinking of it there are many phrase that are just hilarious and have no meaning in other languages... ROFL
03:49:27slaggathoryar
03:49:58 Quit Rick (Read error: 104 (Connection reset by peer))
03:50:40XavierGranother one: do mice -> show me your muscles (in your arms)!!! :}
03:51:20slaggathorah but that is from latin, mus musculus...
03:51:22XavierGrthough this ecaxt word translation so that is making it odd.
03:51:56XavierGrwell mice is a metaphor for muscles
03:52:28TiMiDfrsore wa sou desu ne :)
03:52:37slaggathorso da na
03:52:48XavierGrexcuse me?
03:52:51TiMiDfr:)
03:53:04TiMiDfrnihongo da yo
03:53:11XavierGrchinfe or something?
03:53:17 Join Rick [0] (i=rick@pool-71-108-9-40.lsanca.dsl-w.verizon.net)
03:53:18TiMiDfrjapanese
03:53:25XavierGr^chinese
03:53:28 Quit Bagder (Read error: 60 (Operation timed out))
03:54:33TiMiDfrnihongo ga sukoshi joozu hanasu
03:55:02XavierGroh god it is 5.00 am I will have to sleep, though it was a nice and funny chat!
03:55:08slaggathornihongo no hansu koto ga joozu dewa nai
03:55:14XavierGrgood night
03:55:25slaggathor'nite!
03:55:26 Nick XavierGr is now known as Xavier|Away (n=XavierGr@ppp15-adsl-30.ath.forthnet.gr)
03:55:27TiMiDfroyasumi nasai XavierGr !
03:55:43Xavier|Awayehhmm you too....
03:55:50TiMiDfr:)
03:56:06slaggathors'pose I should go too... thanks again for the remote work, I am using it right now and love it...!
03:56:19TiMiDfrima mo watashi wa
03:56:29slaggathordewa dewa
03:56:36TiMiDfr(4am here and I start class at 9)
03:56:43slaggathorugh
03:56:55slaggathorbonne chance, or some junk
03:56:57TiMiDfr(I will have to sleep)
03:57:02slaggathor^_^
03:57:07TiMiDfryeah bonne chance is correct
03:57:12TiMiDfr1 point :)
03:57:19slaggathorit-ten!
03:57:32TiMiDfrchuugokugo desu ka
03:57:50TiMiDfrouch
03:57:56TiMiDfrichi :)
03:57:57slaggathorchigauze, nihongo da. itten == wan-pointo
03:57:58TiMiDfrok
03:58:06slaggathorhai
03:58:17slaggathoroyasumi
03:58:23TiMiDfrtotemo nemui
03:58:33slaggathorkore mo
03:58:45 Nick TiMiDfr is now known as TiMiD[zzz] (n=TiMiD[FD@asgard.valombre.net)
03:58:45DBUGEnqueued KICK TiMiD[zzz]
03:58:55TiMiD[zzz]dochira he iku ka
03:58:57slaggathormata na
03:59:29TiMiD[zzz]s/iku/kuru
03:59:29 Quit slaggathor ("CGI:IRC (EOF)")
03:59:43TiMiD[zzz]well not an hour for japanese:)
04:00
04:00:07TiMiD[zzz]ja mata ashita
04:00:10TiMiD[zzz]tabun
04:05:46 Join QT [0] (i=as@madwifi/users/area51)
04:14:54 Quit bagawk ("Leaving")
04:15:54 Quit QT_ (Read error: 113 (No route to host))
05:00
05:14:21 Quit paugh (Read error: 113 (No route to host))
05:16:37***Saving seen data "./dancer.seen"
05:26:41 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
05:41:50 Quit Seed (Nick collision from services.)
05:41:57 Join Seed [0] (i=ben@l192-117-115-168.broadband.actcom.net.il)
05:42:22 Join banan_ [0] (i=banan@dalink.campus.luth.se)
06:00
06:00:28 Quit merbanan (Read error: 110 (Connection timed out))
06:43:18 Join webguest67 [0] (n=d86bc594@labb.contactor.se)
06:43:54webguest67hi all
06:45:34webguest67amiconn?
06:46:24webguest67bye
06:46:25 Quit webguest67 (Client Quit)
07:00
07:04:13 Join Birdfish [0] (n=bradbox8@64.108.5.134)
07:05:12 Quit pheon ("Trillian (http://www.ceruleanstudios.com")
07:16:23 Join _FireFly_ [0] (n=FireFly@p54A46A0D.dip.t-dialin.net)
07:16:31_FireFly_hi
07:16:40***Saving seen data "./dancer.seen"
07:25:24Bgernirbu
07:25:28Bgerops, morning :)
07:25:39Bger(fingers disposition)
07:25:47_FireFly_;)
07:32:40_FireFly_bye
07:32:44 Quit _FireFly_ ("Leaving")
07:35:15 Join Gibbed [0] (i=rick@pool-71-108-9-40.lsanca.dsl-w.verizon.net)
07:35:16 Quit Rick (Nick collision from services.)
07:35:26 Nick Gibbed is now known as Rick (i=rick@pool-71-108-9-40.lsanca.dsl-w.verizon.net)
07:44:14Bgerlot of french last night :)
08:00
08:37:12 Join ender` [0] (i=ychat@84.52.165.220)
08:45:38 Join _Vladoman [0] (n=Vladoman@p54A7FB7B.dip.t-dialin.net)
08:57:59 Join LinusN [0] (n=linus@labb.contactor.se)
08:58:43Bgermorning, LinusN
08:58:50LinusNmorning
08:59:54amiconnCowabunga!
09:00
09:00:09LinusNbeaucoup de francais dans les logs
09:01:31 Quit solexx_ (Read error: 110 (Connection timed out))
09:02:38amiconnIt's a pity that logbot still cannot explain anything...
09:03:03LinusNhuh?
09:03:48***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
09:04:56 Quit Vlad0man (Read error: 110 (Connection timed out))
09:07:43amiconnI remember a time when logbot could explain some terms, like LinusN, Bagder, Zagor, quelsaruk, cowabunga
09:07:55amiconn..by using /msg logbot explain <term>
09:08:18LinusNyes, logbot can't /msg anymore, since the nick isn't registered
09:08:55amiconnNo, that's not the problem, I've set unfiltered to on, and logbot answers to my /msg s
09:09:33amiconnIirc the eplain 'list' went away when you upgraded the box some months ago
09:10:06amiconnThe only thing I get from '/msg logbot explain *' (which should bring up the list of known terms is:
09:10:11amiconn[09:02:34] -logbot- *Shrug* I don't know
09:10:52LinusNah i see now
09:11:16LinusNthe explain file has been replaced
09:13:32LinusNpity inideed
09:15:15amiconnIs the old explain file still available somewhere, perhaps an old tape backup?
09:16:43***Saving seen data "./dancer.seen"
09:18:59 Quit bluebrother^ (Read error: 104 (Connection reset by peer))
09:19:23 Quit cYmen (Read error: 104 (Connection reset by peer))
09:19:27 Join cYmen [0] (n=cymen@nat-ph3-wh.rz.uni-karlsruhe.de)
09:19:28 Join bluebrother^ [0] (n=c28@nat-ph3-wh.rz.uni-karlsruhe.de)
09:22:29LinusNamiconn: i'm afraid not...
09:23:00LinusNi'll ask bjrn
09:23:10Bgeramiconn if you have old logs ...
09:24:32Bgersorry for the stupid q, but what are the files that i must edit (except Makefile) to make a new plugin to compile (the plugin is in a new subfolder)
09:24:35amiconnLinusN: Some of the explanations should be in the irc logs, in case someone called them and posted them to the channel
09:24:48LinusNyeah
09:25:13amiconnI can't search across all of them, 'cause google doesn't index them completely
09:25:50amiconnI could download all logs...
09:26:14 Join linuxstb_ [0] (n=5343d4aa@labb.contactor.se)
09:26:15LinusNnot my idea of fun...
09:26:56linuxstb_Bger: You don't need to edit any Makefile - just add the name of the .c file to apps/plugins/SOURCES
09:27:31LinusNlinuxstb_: not if it is in a subfolder
09:27:34Bgerlinuxstb_ the plugin is in new subfolder of plugins
09:29:37linuxstb_Sorry, I've missed that part.
09:30:03LinusNit should be enough to add it in plugins/Makefile
09:30:03Bger10x anyway
09:30:25LinusNBger: look for the SUBDIRS assigment in plugins/Makefile
09:30:33Bgeri copied the Makefile from searchengine's directory and edited it..
09:30:41Bger10x
09:31:47Bgerfound it, trying
09:32:05linuxstb_Bger; What plugin are you working on?
09:32:24Bgerlinuxstb_ don't ask :))
09:33:04Bgerseriously, i'm trying to make encryption plugin for sensitive data
09:33:09Bgerbut i'm in the beginning of it
09:33:34linuxstb_Yes, I remember from yesterday.
09:33:47CtcpIgnored 2 channel CTCP requests in 43 minutes and 59 seconds at the last flood
09:33:47*linuxstb_ drinks more coffee
09:34:04*amiconn found the DownThemAll Firefox extension :)
09:34:20Bgerhehe
09:34:30 Quit linuxstb_ ("CGI:IRC")
09:34:40Bgerbtw FlashGot (if you use it) has such function, iirc
09:36:06 Join solexx [0] (n=jrschulz@c158163.adsl.hansenet.de)
09:36:22Bgerwow, it works ("Hello, world")
09:36:32 Join linuxstb_ [0] (n=5343d4aa@labb.contactor.se)
09:42:35LinusNamiconn: i can prepare a tar file if you're serious
09:42:50Bgerwhile on stupid questions: I have 3 very little funcs() in one header file (byteorder.h), which is included from 2 other C files (aes.c, test.c). is there anything I can do to compile test.o except of (1) writing these 3 funcs in other .C file; and (2) making them "static" (which of course makes them to exist 2 times in the code)
09:42:52amiconnAlready got all logs :)
09:42:57LinusNa biiiiiiig tar file :-)
09:43:06amiconn1207 files up to yesterday
09:43:11Bgerwow :))
09:43:16LinusNenjoy
09:43:18Bgeramiconn - log hunter
09:43:31amiconnIsn't that much data after all, ~64 MB
09:44:11LinusNtrue
09:44:59Bgers/writing/moving
09:49:14 Join tvelocity [0] (n=tony@ipa168.0.tellas.gr)
09:49:58linuxstb_Bger: You could possibly define them as inline.
09:50:11Bgerthey ARE inline
09:50:24Bgerbut ld fails
09:51:00Bgeri'm making gcc test.c aes.c -o test.o
09:52:27Bgeraes.c:(.text+0x0): multiple definition of `rol32'
09:52:27Bgeraes.c:(.text+0xd): multiple definition of `ror32'
09:53:34linuxstb_I would try making them static.
09:53:48Bgeryes, that's what did
09:54:02linuxstb_If they are inline, then it shouldn't make any difference to the code size.
09:54:06Bgerbut this way they are 2 times in the executable
09:54:18Bgeroh, u're right...
09:54:50Bger10x again
09:55:05Bgerdidn't think about this
09:56:14 Nick Xavier|Away is now known as XavierGr (n=XavierGr@ppp15-adsl-30.ath.forthnet.gr)
09:56:25Bgermorning, XavierGr
09:56:30XavierGrHi all and good morning to everybody!
09:56:56 Join B4gder [0] (n=daniel@static-213-115-255-230.sme.bredbandsbolaget.se)
09:57:48Bgerif you remember the phrase "eating wood" today... in bulgarian we have something like "u'll gnaw the wood"
09:58:06XavierGrHaha!
09:58:42XavierGrThen it must be that way with phrase with all languages...
09:58:42XavierGr^phrases
09:59:16Bgerbut it has different meaning ... like "u'll hug the bouqet" which is equal to "u'll die" ...
10:00
10:00:03XavierGror ... you will see the flowers upside down!
10:00:05Bgeror "u'll kick the bell" ... better stop:)
10:00:13XavierGr(from underneath)
10:00:24Bgeryea
10:00:27XavierGr:D
10:00:54linuxstb_Bger: How is your plugin going to work? What is the aes routine going to decrypt?
10:01:02linuxstb_A text file?
10:01:27Bgeri'm thinking of it like Cipher=AES(K,Plaintext)
10:01:36Bgerwhere K=SHA256(Password)
10:01:56linuxstb_I'm wondering what it will do from a user's point of view.
10:02:25linuxstb_i.e. How will the data be entered and encrypted in the first place?
10:03:02Bgeri'm thinking about it yet
10:04:51Bgerit'll be good to be editable on both the player and the PC
10:05:55linuxstb_bbl.
10:05:58 Quit linuxstb_ ("CGI:IRC")
10:09:02Bgeris the "parameter" in plugin_start() the filename when it's started as a viewer ?
10:10:18B4gderdtl
10:10:21B4gdererrr
10:10:28B4gderdotcom-Zagor is in all media today
10:10:34B4gderSwedish that is
10:12:11BgerB4gder did u read about my stupid problem yesterday ?
10:12:25B4gderno, I missed that
10:12:45B4gderwhat was it?
10:12:59Bgerthe one of the aes_setkey routines takes the keysize in bits, the other - in bytes ...
10:13:10B4gderaaah... hehe
10:13:19BgerAES's keysizes are 128,192 and 256 bits
10:13:25Bgeriirc
10:14:18Bgerhm, i see the best format for such plugin's input like the dictionary plugin
10:14:59Bgerlike passwordname passwordvalue
10:15:00Bgeretc
10:15:38Bgerand if it's not in this format, show it as text
10:17:06Bger(not editable)
10:20:44 Join pilot000 [0] (n=51dd1b21@labb.contactor.se)
10:22:42 Quit cYmen (Read error: 104 (Connection reset by peer))
10:22:43 Quit bluebrother^ (Read error: 104 (Connection reset by peer))
10:26:04 Join cYmen [0] (n=cymen@nat-ph3-wh.rz.uni-karlsruhe.de)
10:26:40 Join bluebrother^ [0] (n=c28@nat-ph3-wh.rz.uni-karlsruhe.de)
10:28:31 Join preglow [0] (n=thomjoha@hekta.edt.aft.hist.no)
10:32:00 Join ashridah [0] (i=ashridah@220-253-120-63.VIC.netspace.net.au)
10:38:44preglowdotcom-Zagor is zagor's evil twin? :>
10:39:18B4gderyeah
10:39:35preglowwhat's he up to?
10:39:58B4gderhe sold his site to a major player on the market
10:40:12LinusNfor some major money :-)
10:40:16B4gderbjornsbytare.se
10:40:19preglowhahah
10:40:33preglowthis means he'll have more rockbox time, yes? :P
10:40:40B4gdernonono, we jealous? ;-]
10:40:48LinusNnaaaaaaa
10:45:48preglowright, he'll continue working with it
10:46:02B4gderyes, they hired him too
10:48:42Bgerthe only thing i understand on this site is "annons", "forum" "logga in" "diskutera i forumet" "statistik" "lista alla" and "programmering" :))
10:48:53B4gderhehe
10:49:04B4gderBger: its a site for swapping appartments
10:49:13Bgeraha :)
10:49:17B4gderrental appartments, in Sweden
10:49:44Bgeranyway, i suppose this is good news :)
10:50:13B4gderit proves that many years of spare time efforts can pay off in the end...
10:50:29Bger:)))
10:50:51preglowdoesn't work for me :/
10:51:02Bgerhm, i forgot the stamp "originalet"
11:00
11:08:37 Join linuxstb [0] (n=linuxstb@213.86.218.27)
11:16:45***Saving seen data "./dancer.seen"
11:17:10 Quit markun (Read error: 60 (Operation timed out))
11:28:01 Join markun [0] (n=karl@bastards.student.ipv6.utwente.nl)
12:00
12:09:35 Part LinusN
12:24:11 Join LinusN [0] (n=linus@labb.contactor.se)
12:31:33 Join Febs [0] (n=Febs@207-172-122-81.c3-0.rdl-ubr4.trpr-rdl.pa.cable.rcn.com)
12:36:27LinusNSlasheri: u there?
12:37:17SlasheriLinusN: hi, yes :) but little busy, at work now
12:37:41LinusNi have tried your cache patch
12:37:55LinusNworks fine
12:37:59Slasheriah, that's good :) what do you think about it?
12:38:03Slasherigreat :)
12:38:14preglowworks fine for me as well
12:38:24LinusNone nitpick: i'd like some indication of what's going on when building the cache for the very first time
12:38:31Slasheriso it should be ready to commit it and start doing bugfixes etc?
12:38:49LinusNright now it stays with the logo and doesn't say anything
12:38:52 Join Moos [0] (i=DrMoos@m62.net81-66-158.noos.fr)
12:39:02SlasheriHmm, good idea. Could it be possible to write something to the screen using lcd_puts or similar?
12:39:12LinusNsure
12:39:23Slasheriok, i will implement that :)
12:39:40LinusNso, what to do with the fastboot?
12:40:06LinusNi think the fastboot should be removed
12:40:18Slasherii left the option in the code but ifdeffed it out so it could be enabled in future if necessary
12:40:26XavierGrfastboot?
12:40:26LinusNand if we really want it, we could add it later
12:40:27preglowfastboot?
12:40:31Slasherido you think it could be better to completely remove the code?
12:40:37Slasheriok
12:40:47LinusNyes, i don't like dead code
12:41:06preglowwhat is it?
12:41:13Slasherii will remove it then (and the load and save functions too?)
12:41:17LinusNpreglow: saving the cache in a file
12:41:20Slasheripreglow: cache file
12:41:28LinusNpreglow: and load it at boot instead of scanning
12:41:34preglowwhat's wrong with that?
12:41:42LinusNconsistency problems
12:41:44XavierGrwhy not have the cache file?
12:41:55SlasheriLinusN: or what if i just #if 0 the load and save code (two functions)? And remove the options completely
12:41:56XavierGrlike?
12:42:01preglowright, so there's always a blocking scan at the start?
12:42:38LinusNpreglow: no
12:42:41Slasheripreglow: it's always a transparent scan (except the first time)
12:43:12XavierGrwhisch is the duration of a fast boot comapring a no cache file boot Slasheri?
12:43:14preglowso, the first scan is now just about finding the cache size?
12:43:27LinusNpreglow: more or less
12:43:33preglowok, i can live with that
12:43:48SlasheriXavierGr: the boot time is save no matter if cache (or fastboot) is enabled
12:43:49LinusNSlasheri: #if 0 is dead code too :-)
12:43:51Slasheri*same
12:43:53XavierGrany major difference in timing?
12:43:59SlasheriLinusN: hehe, true :D
12:44:15XavierGrok then I can live with that too.
12:44:43SlasheriLinusN: but anyway, i will remove the code from the options and consider what to do with the save & load
12:45:10XavierGrLinus: Did you find time to check fm preset and/or jpeg viewer?
12:45:23LinusNhmmm, the save & load might still be nice for debugging
12:45:37LinusNXavierGr: that's next on my agenda
12:46:07XavierGr:D thanks I was a little worried...
12:46:23SlasheriLinusN: yes, it's possible the functions might be useful at some situations. So maybe i will let them be there some time so those can be removed later
12:46:33LinusNdo so
12:47:11*preglow does the commit dance
12:47:39B4gdermy favourite!
12:48:04LinusNXavierGr: which patch is it?
12:50:29 Join resiak [0] (n=will@unaffiliated/resiak)
12:51:43XavierGrIn the patch tracker you can find both the fm preset patch and the jpeg filescrolling patch.
12:52:01XavierGrwai a sec to get the links...
12:52:21XavierGrhttp://sourceforge.net/tracker/index.php?func=detail&atid=439120&group_id=44306&aid=1312132
12:52:26XavierGrfm preset patch
12:52:32LinusNfound it, thx
12:52:35XavierGrhttp://sourceforge.net/tracker/index.php?func=detail&atid=439120&group_id=44306&aid=1266294
12:52:42XavierGrJPEG file scrolling patch
12:53:03XavierGryou're welcome! :)
12:55:36LinusNcan you think of an icon for the fmr files?
12:56:16XavierGrah yes I forgot that, I put the same as for playlists.
12:56:25XavierGrbut are there any vaccant?
12:56:39XavierGrI don't think I can draw one.
12:57:40XavierGrI think someone here had a gallery with little Rockbox icons...
12:58:24preglowwww.rasher.dk/rockbox/
12:58:27XavierGrwe could use something with a note and the little "stripes" of the playlist icon.
13:00
13:00:18XavierGror even an icon with FM written on it.
13:01:01 Quit paugh ("Leaving")
13:01:17XavierGroh this page has the codes too so it will be easy to draw one.
13:06:17XavierGr2 fit 2 letters seems difficult.
13:06:47 Quit tvelocity (Read error: 110 (Connection timed out))
13:12:05 Join tvelocity [0] (n=tony@ipa205.4.tellas.gr)
13:12:21preglowif not impossible
13:12:33XavierGrhi tvelocity!
13:13:03XavierGrhmm I am a lousy drawer even in this I can't make a good icon.
13:13:14XavierGrRasher got some talent!
13:16:46***Saving seen data "./dancer.seen"
13:17:57LinusNXavierGr: i'm curious about the usage of fmradio_get_status()
13:19:37LinusNyou don't use the FMRADIO_xxx defines
13:20:01XavierGrwell I call it to see if the radio is already started. if it is and the user is in filtree we call radio_screen else not.
13:20:54XavierGrwhy use the FMRADIO_xxx defines?
13:21:22LinusNbecause they are the values you should use when getting setting the status
13:22:21LinusNif(radio_get_status() != 1)
13:22:26LinusNshould be
13:22:34XavierGrwait a minute I thought that fmradio_get retunrs an integer?
13:22:34LinusNif(radio_get_status() != FMRADIO_POWERED)
13:22:43XavierGrah you mean to make it clear
13:23:01LinusNof course
13:23:03XavierGryeah that is more elegant
13:23:25LinusNand future proof, if we add more states
13:23:40LinusNkind of the whole point of using #defines
13:23:43XavierGrwell only 2 calls in radio.c and 1 in filetree.c
13:24:30XavierGryes you right, I am not using in a so defined programming(well I am not used in programming itslef but that another story)
13:25:06LinusNin fact, i see an old bug
13:25:41LinusNwe leave it as FMRADIO_POWERED if keep_playing is true, but it should probably be FMRADIO_PLAYING
13:26:12XavierGrbut then I couldn't be sure if the radio is shown
13:26:18LinusNdoesn't really matter
13:26:21XavierGrbetter this way.
13:26:26LinusNtrue
13:26:41LinusNyou're right
13:33:41 Nick _Vladoman is now known as Vladoman (n=Vladoman@p54A7FB7B.dip.t-dialin.net)
13:46:29LinusNXavierGr: you should try to follow the CONTRIBUTING guidelines regarding the code style
13:46:49LinusNi.e place the braces consistently
13:46:56XavierGrI think I mostly follow it what did I forget?
13:47:30XavierGrwhy aren't they well fitted?
13:47:41LinusNthe opening brace should be on a separate line in this particular file
13:47:46XavierGrI am moving to the link on how it must be.
13:47:47LinusNwell fitted?
13:48:13XavierGrI mean well placed in order to the contibuting guide.
13:48:29LinusN"When changing code, follow the code style of the file you are editing."
13:48:57LinusN"When writing new files, you may use the brace placement style of your choice."
13:50:01XavierGrI thought that we change line on braces only for functions.
13:50:50XavierGryes you are right in radio.c all ifs and whiles have their brackets underneath.
13:50:57XavierGrmy bad.
13:51:02LinusNnp
14:00
14:00:36 Quit Febs (Read error: 110 (Connection timed out))
14:01:06LinusNXavierGr: it seems that there is no default preset file anymore
14:01:23LinusNso if i start the fm radio and add a preset, it says "Preset save failed"
14:02:20novimondoes anyone know why rockbox freezes when I try to play flac?
14:02:35LinusNi didn't even know it did
14:02:45XavierGrno it shouldnt. it checks if the directory .rockbox/presets exitst if not it will create it.
14:02:50XavierGrI will test again.
14:03:42XavierGrdammit you are right.
14:03:59LinusNfilepreset[] isn't initialized anymore
14:04:01preglownovimon: you tried lots of files?
14:04:49Bgerhttp://muppets.go.com/games/muppets-flyinggonzo.swf
14:05:24XavierGrLinus: yes my bad I have put the code that checks for the folder in save_preset list which calls save_presets.
14:05:59XavierGrso the very first time that you add a preset it doesn't call the save_preset_list so there will be no folder creation
14:06:16XavierGrbut when you add a preset the save_presets is called it can find the folder and fails.
14:10:29*Bger wants flash player on rb :)
14:10:30XavierGrI found a solution.
14:10:48LinusNBger: :-)
14:11:13Bgeror at least Java ME
14:11:16Bger:)
14:11:56preglowgood luck on that
14:12:13XavierGrhttp://pastebin.com/384868
14:12:32XavierGrLinus you can change the save_presets to this.
14:13:04XavierGrthis is a bug that will only appear the first time. Or if there is no presets directory.
14:16:16XavierGrso much for my debugging. and I thought that I had find all strange posibilities
14:16:56LinusN:-)
14:17:54preglowSlasheri: so, when do you expect to have dircache commited?
14:18:03LinusNXavierGr: what's the purpose of this:
14:18:09LinusN if(strcmp(preset_file, "/"))
14:18:15XavierGrof this change?
14:18:25preglownovimon: i can play flac with no problems here
14:18:28LinusNno, the test itself
14:18:31XavierGroh yes I have
14:18:51XavierGrput an option clear preset list
14:19:03XavierGrwhich calls the load preset list as (/)
14:19:55LinusNi don't get it
14:20:08XavierGrit will refuse to save an empty list
14:20:29LinusNbtw, you use mkdir(PRESET_PATH, 0), but not in the creat() call further down
14:20:33XavierGrif the user clears the list and then tries to save it without adding a preset first.
14:21:01LinusNhow about num_presets == 0?
14:21:08XavierGrthe second time I just put the whole filename for a generic preset list.
14:21:54LinusNcreat("/.rockbox/presets/deault-fm.fmr", O_WRONLY);
14:22:20LinusNi think presetfile[] should be preinitialized to this
14:22:32LinusN(and you have misspelled "default")
14:22:58XavierGrgreat! I just rushing things in order to make the bugs go away....
14:23:06XavierGrwait a sec
14:24:01XavierGrnum_presets == 0 prevents the rendering of an empty list if the user push play while there is no presets.
14:24:55LinusNyes?
14:25:15LinusNnum_presets == 0 means "the preset list is empty"
14:25:47LinusNthe same should go for radio_save_presets()
14:25:58LinusNor am i missing something?
14:26:42Slasheripreglow: this week (maybe weekend)
14:26:57XavierGryes so I set presets_loaded to false in order NOT to run handle_radio_presets()
14:27:40XavierGrradio_save_presets only appear when someone manually adds a preset or save the preset list.
14:27:53XavierGrin both options i think there is no need to check it
14:28:07XavierGrif the user adds a preset num_preset will no longer be 0
14:29:34XavierGrno I am wrong if the user tries to save an empty list it will not stop him
14:29:59LinusNin fact, presets_loaded is unnecessary
14:30:05XavierGrso yes we need to put that if(num_preset == 0) in save_preset_list too
14:30:18XavierGrI need that for a special case in filetree.c
14:30:23XavierGrthat's why it is global.
14:30:41XavierGrit will check if a preset outside the presets folder is loaded.
14:31:27XavierGrif I remove it even if you open a preset file outside the default folder the "settings" preset file will be loaded.
14:32:51LinusNbut that's the filepreset_load flag, not presets_loaded
14:33:10XavierGroh yes sorry
14:33:33XavierGrpresets_loaded is an old bool that I didn't make
14:33:39LinusNyes
14:34:13XavierGrif not needed we can put it out. I was just afraid that a bug will occur.
14:34:17Bgerbtw what does "warning: creating a DT_TEXTREL in object" mean ?
14:34:43XavierGrhttp://pastebin.com/384892
14:35:02XavierGrthat is for the save_preset_list
14:35:10XavierGrI should compile it now to see.
14:36:34 Join Zagor [0] (n=bjst@194-237-150-170.customer.telia.com)
14:36:56preglowdotcom-Zagor!
14:37:05Zagorhehe, yeah
14:37:13linuxstbnovimon: Do your FLAC files have ID3 tags?
14:37:15LinusNXavierGr: take your time and produce an updated patch
14:37:37preglowid3 tags on flac files, when will the madness end...
14:37:45B4gderZagor: you even reached DN Ekonomi
14:38:05XavierGrLinus: is there anything else you want me to change?
14:38:08ZagorB4gder: ooh. paper or just net?
14:38:14B4gderpaper
14:38:17XavierGrTest this patch in a biuld to see if you can find other annoyances.
14:38:19B4gderdidn't check online
14:38:37Bgercongrats, Zagor
14:38:46linuxstbpreglow: I've started extracting the ffmpeg flac decoder. I got as far as compiling it as a standalone decoder (on my PC) last night, and will try and make it actually decode files thiis evening.
14:39:15linuxstbIt looks very promising.
14:39:31preglowlinuxstb: remarkable
14:39:57XavierGrLinusN: Also you can check the jpeg patch too if you have time, so if you find something odd tell me to change it too.
14:40:06crwlhow's the ffmpeg flac decoder better than the current one?
14:40:20linuxstbcrwl: It's several orders of magnitude simpler.
14:40:38linuxstbWhich means it is a lot easier to optimise for Rockbox.
14:40:51preglowi wonder if it differs in functionality
14:41:09linuxstbI've no idea yet.
14:41:26crwlok.
14:42:45preglowdid you have to extract many other ffmpeg components to make it work standalone?
14:43:15linuxstbA little, but not too much. Mainly just the bitreading functions and other small utility functions.
14:44:03novimonlinuxstb, yeah, they have
14:44:29novimonpreglow, and yes, i've tried three albums
14:45:04preglownovimon: id3 tags in flac files is badness
14:45:05linuxstbnovimon: That's the problem then. Rockbox doesn't support FLAC files corrupted with ID3 tags.
14:45:18preglowbut i expect we'll have to now...
14:45:22linuxstbOut of curiosity, which application created them?
14:45:48novimonpreglow, okay then, can I somehow remove the tags without need to re-encode
14:45:53novimonlinuxstb, dbpoweramp
14:45:56preglownovimon: i have no idea
14:46:00preglowlinuxstb: please mail them and have them stop :>
14:47:11linuxstbDear Mr. D.B. Poweramp. Stop it. Love Rockbox.
14:47:20B4gderhaha
14:47:30B4gderthat'll do it
14:48:07novimonremoved the tags now
14:48:13novimonlet me test..
14:48:39novimonanyone modded the senn hd414 pads to alessandro ms-1s?
14:50:56preglowDear Mr. Amp. This is folly. Hugs and kisses from the Rockbox team.
14:51:17preglowi thought i saw some id3 handling code in libflac some place
14:51:57novimonrock box does really rock thou :)
14:52:00markunhttp://lists.xiph.org/pipermail/flac/2005-January/000367.html
14:52:19linuxstbYes, libFLAC should skip ID3 tags in FLAC files, but the code in metadata.c doesn't. There may also be other problems - I haven't looked at the issue.
14:52:34novimonis there any point updating dayly?
14:52:54novimonis there a big possibility that I break something if I do? :D
14:53:01linuxstbCheck the CVS changes (on the Rockbox home page) - if a new feature or bugfix that you want has been implemented, then update.
14:53:22novimonokay
14:53:34linuxstbIt is very rare for a new version to be worse than a previous version. If that happens, it is normally fixed quite quickly.
14:53:38novimonthat I've done so far
14:53:43novimonokay
14:53:52novimonso its safe to be up-to-date :)
14:54:44XavierGrokay launch time. Linus I will send you later the patch.
14:55:36novimondoes btw rockbox support the replaygain value in the flac?
14:55:45crwlyes
14:56:36novimonok, and does flac have it's own metadata?
14:57:37MoosLinus: thanks for the solitaire patch :)
14:58:05crwlnovimon, yes
14:58:36novimonhmm, gotta pressure to get the dbpoweramp team to start using it :P
14:59:07B4gdernovimon: tell them they don't get any bonus points by us otherwise! ;-)
14:59:27linuxstbYes, FLAC uses Vorbis Comments as its native tag format - the same format as Ogg.
14:59:35novimonyeah, need to :)
14:59:56novimonlinuxstb, dbpoweramp makes ogg "tags" nicely thou
15:00
15:00:01novimonhmm
15:00:21novimonmaybe it's a mater of configuration, or then not
15:00:33linuxstbIs there a checkbox to enable/disable ID3 tags?
15:00:46novimonlinuxstb, yeah
15:01:05linuxstbThen I guess you should disable them when creating FLACs
15:01:18novimonits not coded depended thou, so if I want to make oggs I'd rather keep it on..
15:01:23novimon*codec
15:01:37 Part LinusN
15:01:40B4gderid3 on oggs?
15:01:42linuxstbYou don't want ID3 for Oggs either.
15:02:05fuzzieyou don't want id3 on anything except mp3, do you?
15:08:16novimonw00t, it works
15:08:18novimonthanks mates :)
15:08:46 Join webguest88 [0] (n=d4406110@labb.contactor.se)
15:09:25novimonfuzzie, I think it means ID3 as metadata generally
15:10:11webguest88would anyone be interested commiting the patches for wps replay gain track and album, also crossfade wps tags
15:11:40 Part webguest88
15:13:44solexxwhy no id3 on ogg?
15:14:11preglowbecause id3 doesn't belong on oggs
15:14:17preglowogg has got it's own metadata format
15:14:26B4gderand id3 is especially made for mp3
15:14:57solexxwhere's the difference?
15:15:11B4gderthey're very different, formatwise
15:15:18solexx(iow: why should I care?)
15:16:49***Saving seen data "./dancer.seen"
15:17:49ashridahbecause you asked why id3 isn't used on ogg.
15:18:14ashridahthis could also be considered as 'it makes no difference to a user'
15:18:51solexxhm, ok.
15:20:07solexxis vorbiscomment supported by rocvkbox at all?
15:20:11preglowyes
15:20:13preglowdvs
15:20:19preglowis vorbiscomment the tag spec? :P
15:21:07solexxdvs? ("Domestic violence services" is what google proposes :))
15:21:44preglowsorry, norwegian abbreviation, i meant "that is"
15:21:45preglowheh
15:23:17preglowusing id3 files on ogg files is like using vorbiscomment on mp3 files
15:24:01solexxok, i got it.
15:24:02preglowand it blatantly ignores the fact that an ogg file isn't analogous to an mp3 file at all, one ogg file can contain lots of data, not just an audio track
15:25:04solexxhm. is there a simple way to change id3 tags to vorbis comments?
15:25:33solexxbest thing would be sth like vorbisgain, which recursively descends into subdirectories
15:27:03preglowno idea
15:27:09solexxhm, the more i think of it, the louder i hear the word "python" on thy back of my head
15:27:20preglowgoing insane? :P
15:27:48solexxi wish i would! :)
15:28:22preglowyes, that would fit me like a glove
15:30:10B4gderwww.rockbox.org had 60000+ unique visitors during September
15:32:06B4gderhttp://lwn.net/Articles/151790/
15:33:43 Join Paul_The_Nerd [0] (n=paulthen@cpe-66-68-93-2.austin.res.rr.com)
15:33:46solexxw00t!
15:34:02Paul_The_NerdXavierGR: Are you present?
15:34:50B4gdernice little piece there about our project
15:36:26XavierGryes
15:36:51preglowit's even accurate
15:37:04Paul_The_NerdDid you ever get an icon for FM playlists?
15:37:30Paul_The_NerdI've mocked a simple FM one up in that 6*7 format from rasher.dk
15:38:58*B4gder reads a public comment from a guy at irivernordic.com
15:39:27B4gdersaying: "Rockbox to h100 is very good. Use it on your own risk, we can't be responsible for what happens"
15:39:32B4gderrockbox on
15:39:42B4gder(translated from swedish be me)
15:40:25preglowhaha
15:41:12Paul_The_NerdAlso, as a note for the logs, I have *no* idea how novimon got dbpoweramp to use ID3 tags, because it's always used native tagging formats for any file I've used it to create, including FLAC.
15:41:55XavierGrPaul_The_Nerd: Yes I would like to see that.
15:42:38novimonPaul_The_Nerd, I just simply checked the "use id3tags" checkbox :)
15:42:56Paul_The_NerdUsing the audio cd input?
15:43:33Paul_The_Nerdnovimon: Are you using the newest version, or an older one, as mine has "Write id tags"
15:44:44Paul_The_NerdIt's tag format agnostic and seems to use the appropriate type for the file.
15:44:52solexxcool, i just realize that tagtool (my tagging tool of choice) uses vorbiscomment...
15:45:37solexxif i only had tagged my whole collection with it...
15:46:06solexxno python for me today
15:47:58XavierGrPaul_The_Nerd: ok thanks i have to go now. later
15:48:09Bgerhow much is one tick ?
15:49:15amiconnTiMiD[zzz]: There?
15:49:23Bgerother way: what does sleep(100) mean ?
15:52:54B4gderBger: HZ is one second
15:52:58linuxstbBger: Time in Rockbox is measured by HZ.
15:53:38novimonPaul_The_Nerd, I have the newest one, didnt what is was called
15:53:39linuxstbe.g. sleep(HZ/2) sleeps for half a second.
15:53:58Bger10x, i already realised this
15:54:02novimonPaul_The_Nerd, and yes, using audio cd input
15:54:16Paul_The_NerdNovimon: Where is the id3tags checkbox then?
15:54:53novimonoptions -> write id3 tags
15:56:42novimonhaa, found a track creation metadata option
15:56:59novimonput it to vorbis comment
15:57:30Paul_The_NerdMy options window doesn't have write id3 tags, it simply has write ID tags.
15:57:54 Quit pilot000 ("CGI:IRC")
15:58:41novimonmine has only "write ID tags" too
15:58:48novimonread it too fast I suppose :)
15:58:51Paul_The_NerdAh
15:59:12Paul_The_NerdAnd then under the more advanced configuration, there's a "ID Tag options" where you can set the default tags.
15:59:20novimonyeah, found that
15:59:34Paul_The_NerdMine defaulted to Vorbis Comments for FLAC though.
16:00
16:00:40novimonit could be that I have earlier changed ti and then forgot it :)
16:01:46Paul_The_NerdHeh
16:05:51Bgeramiconn how do i clear the display ?
16:06:12 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
16:06:34Bgerfound it
16:13:53*Bger searchs for experimental "mouse"
16:14:14Bgerseriously, someone with iriver h120 here ?
16:14:40Bgeror h140
16:18:42Bgeri just want to test something on the target ...
16:19:01Bgermore exactly sha256
16:19:20linuxstbI can test something in about 3-4 hours time, but not now.
16:19:28Bgerokay
16:21:16Bgerhm, i'll make some more changes
16:22:12Bgercan i pass NULL as param to lcd_getstringsize()?
16:24:36B4gderyes
16:25:00Bger10x
16:25:15*Bger is "Mr. stupid questions" last 2-3 days
16:25:35B4gderno worries
16:25:47B4gderwe'll shutup when we get tired ;-)
16:26:56Bgerhehe ok
16:27:21Bgermost of u already shuted up:)
16:27:28B4gderhahaha
16:29:20 Join Raxus [0] (i=Raxus@athe530-f019.otenet.gr)
16:31:54BgerB4gder do u have opportunity to test a .rock ?
16:32:14B4gderhm. sure, I'll try
16:32:22Bgeron iriver
16:32:41Bgermy first rockbox code:P
16:33:02B4gderhit me!
16:33:07Bgere-mail ?
16:33:11Bger(i know it)
16:33:13B4gdersure
16:33:19Bgersec
16:33:36Bgerhm, i suppose it'll be very slow ...
16:33:47Bgermaybe i can add a timer also
16:33:53 Join Lost-ash [0] (i=ashridah@220-253-122-177.VIC.netspace.net.au)
16:34:23 Quit ashridah (Nick collision from services.)
16:34:27 Nick Lost-ash is now known as ashridah (i=ashridah@220-253-122-177.VIC.netspace.net.au)
16:35:04MoosBger: what does it for?
16:35:10Bgeri suppose the tm struct is like linux's
16:35:20Moospassword thing?
16:35:21BgerMoos it doesn't do any usable :)
16:35:23Bgeryep
16:35:38Mooscool :-)
16:36:04Bgerit's too early to say it's cool :)
16:36:18Mooshehe :)
16:36:40Moosit's cool that you'r working on it :D
16:37:33Bgeri suppose get_time is not working on iriver
16:37:41Bgerheh, 10x :)
16:39:51BgerB4gder heeeelp... how can i measure diff in time in rockbox ?
16:43:42B4gder*rb->current_tick perhaps?
16:44:53Bger:))
16:45:13Bgerhm sha256 maybe will not be so slow, but aes ....
16:45:43Bgerthe aes test takes 10-15sec/turn on duron@900MHz
16:45:50B4gderwow
16:46:30*Bger doesn't want to think how much it'll take on coldfire@120Mhz ...
16:47:09ashridahcan you use any of the advanced features of the coldfire to optimise it?
16:47:51Bgerashridah the test is something such as 400 * 9999 encryptions ...
16:47:58Bgerof 128bit blocks
16:47:59B4gderI think you could use a less paranoid encryption
16:48:18Bgermaybe, we'll see
16:48:49B4gdercompare the distributed.net rc5 72 bit compo
16:49:35Bgerthat's about 40*10^9 bytes
16:49:43Bgerhm, error
16:50:34Bger64*10^6
16:50:50ashridahnice
16:51:14amiconnBger: Your aes stuff will be fun on SH1 @11 MHz then...
16:51:19Bgerbtw, i think i use worse algo than i can
16:51:23ashridahcrap. almost 1am. gotta hit the sack
16:51:24 Quit ashridah ("Leaving")
16:52:19Bgersee www.cr0.net:8040/code/crypto/aesbench
16:52:33Bgeri use code based on gladman's one
16:54:06Bgeranyway, changing the algo/encryption is not so difficult
16:54:35Bgeralso, there are many ror32 and rol32 in the code and i'm sure there are such sh1 and coldfire instructions
16:54:39Bgeror i'm wrong
16:55:02amiconnDoes aes use datatypes >32 bits?
16:55:09Bgerno
16:55:17Bgeru32 max
16:55:17amiconnThe SH1 does have rotate, the coldfire does not
16:56:03Bgerhm, worse news... it does SWAB
16:56:08Bgeron big endian
16:59:07Bgeranyway, if it runs, asm heros like u and preglow can take a look at some things
16:59:54amiconnEndian swap is really fast on SH1 (3 asm instructions for 32 bit)
17:00
17:00:46Bgerand on coldfire ?
17:01:25amiconncoldfire has no byte swap, only word swap
17:01:36BgerSWAB32
17:01:47amiconn...but using asm the 32 bit endian swap is still somewhat faster than the C version
17:01:48amiconnyes
17:02:38Bgeri've already #define-d le32_to_cpu and reverse as SWAB32 if ROCKBOX_BIG_ENDIAN is defined
17:08:16 Nick banan_ is now known as merbanan (i=banan@dalink.campus.luth.se)
17:08:25 Quit ender` (Read error: 104 (Connection reset by peer))
17:14:11linuxstbamiconn: The problem is that the SWAB routines are only defined on big-endian targets, and are intended for little-endian to host-endian conversions.
17:15:10linuxstbI would like to make them more general (they don't follow the posix definition of SWAB), and then use macros to define the various endian conversions in terms of those asm-optimised functions.
17:15:41linuxstbBut it touches the low-level FAT/ATA drivers, and I would prefer not to go near there.
17:16:52 Quit Raxus ()
17:16:53***Saving seen data "./dancer.seen"
17:19:43 Quit Moos (Read error: 110 (Connection timed out))
17:23:44amiconnlinuxstb: Are there recommended/standard macro names for converting big endian <-> host endian, and little endian <-> host endian?
17:25:33 Join ender` [0] (i=ychat@84.52.165.220)
17:28:19amiconnlinuxstb: Btw, you will need to touch the ata code anyway for your ipod port...
17:29:02Bgeramiconn can u test my "test" plugin now ?
17:31:24amiconnNo, soryy
17:31:29amiconnlater...
17:31:30Bger:( ok
17:31:55Bgeri made a simple % counter...
17:32:04Bgerso u won't think it has frozen
17:32:44Bgerif it works generally
17:34:47linuxstbamiconn: Yes, I know about my ipod port.... So I may well look at it.
17:35:07linuxstbI don't know what the standard names are for such functions, but I'll try and find out.
17:38:14amiconnI know about htonl() and ntohl(), but these are only for big endian ("network byte order") <-> host endian
17:38:46amiconn...and 32 bit. Are there 16 bit versions??
17:39:00B4gderan idea could be to see what names the linux sources use for this kind of macros
17:39:35BgerB4gder do u have time and can test ?
17:39:48B4gdersend it over and I'll try
17:39:54Bgerjust sec
17:44:03BgerB4gder sent
17:44:23Bgersource + .rock
17:48:27Bgerthe only thing i can say is that it's working in the simulator
17:51:24Bgeramiconn your mail ?
17:51:34Bgeris the mail in the wiki actual ?
17:51:34amiconn16 bit versions of the network byte order fns do exist: htons() and ntohs()
17:51:43amiconnThese four are posix standard
17:54:21amiconnQNX has a set of macros: http://www.qnx.com/developers/docs/qnx_6.1_docs/neutrino/lib_ref/summary.html#ConversionFns
17:55:02 Quit Zagor ("Leaving")
17:56:46Bgergotta go
17:57:21Bgeranyway, B4gder if u test it, write a note in the irc, maybe i'll check the logs later
17:58:01Bgerif u can, make a crude estimation of the time. 10x a lot
18:00
18:00:31amiconnWhy doesn't your plugin time itself?
18:01:51amiconnlinuxstb: Looks like this is what we need to implement: http://rootr.net/man/man/byteorder/3
18:02:22amiconn(probably without the *64 versions)
18:02:44linuxstbI like the bugs listed at the end of the page...
18:03:35amiconnhehe
18:04:02B4gderBger: "I03: AddrErr" after test 2 - only a few seconds
18:04:34linuxstbBut yes, they look like a good standard to adopt.
18:05:34linuxstbSo the asm routines should be called swap16 and swap32, and we define everything as macros in terms of those functions.
18:06:59linuxstbI think we may need functions to swap in-place, as well as functions returning the swapped value. But I'll investigate.
18:07:06amiconnSome of these byteorder things are actually aliases
18:07:21amiconnE.g. htonl() is the same as htobe32(9 etc
18:07:57linuxstbI was thinking about just removing the network-order versions. Just keep it as le/be and 16/32.
18:08:14amiconnThere are 8 different macros, 4 of them are identity macors on either endian architecture
18:08:37 Quit HCl (Remote closed the connection)
18:09:06amiconnYeah, probably the network order version are unnecessary as our targets don't have network interfaces.
18:09:25amiconnHowever, these are posix standard, unlike the le/be, 16/32 versions
18:09:37linuxstbBut it would cost nothing to add the macros, so I may as well add them in at the same time.
18:12:50 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
18:16:48 Join HCl [0] (i=hcl@titania.student.utwente.nl)
18:17:34 Quit B4gder ("time to say moo")
18:18:05 Join jeddy3 [0] (n=jeddy3@bre91-1-82-238-224-40.fbx.proxad.net)
18:18:27jeddy3m
18:21:21 Join _FireFly_ [0] (n=FireFly@p54A4473E.dip.t-dialin.net)
18:23:04 Quit jeddy3 ("Lost terminal")
18:23:44 Join jeddy3 [0] (n=jeddy3@bre91-1-82-238-224-40.fbx.proxad.net)
18:24:47jeddy3exit
18:24:54 Quit jeddy3 (Client Quit)
18:26:33 Join jeddy3 [0] (n=jeddy3@bre91-1-82-238-224-40.fbx.proxad.net)
18:50:21 Part linuxstb ("Leaving")
18:56:01preglowsomeone figure out how to do proper mp3 gapless :/
19:00
19:10:53amiconnpreglow: Huh?
19:16:54***Saving seen data "./dancer.seen"
19:17:01amiconnNiiice - logbot logged the 'expadd' commands :)
19:17:12preglowamiconn: mpa.c isn't proper gapless yet
19:17:21amiconnIt's not?
19:17:28preglowno
19:17:31preglowsmall small gaps left
19:25:38_FireFly_ok id3tag viewer works in the remote :)
19:33:50_FireFly_will someone make the playlist-viewer ;)
19:46:42amiconnexpadd oogabooga=The secret Rockbox salute.
19:46:47amiconnmeh.
19:47:16_FireFly_??
19:47:33*amiconn isn't allowed to re-add that stuff...
19:47:38amiconn:(
19:47:51_FireFly_expadd oogabooga=The secret Rockbox salute.
19:47:54_FireFly_:)
19:48:22SlasheriBger: blowfish should be fast and secure encryption also
19:48:24amiconnThat was a mistake, should have /msg logbot prepended
19:48:41amiconn...but I'm not allowed to use the xpadd command
19:48:46amiconn*expadd even
19:49:02*amiconn found all 15 explanations
19:50:24SlasheriBger: but i don't undestand how you got aes working so slowly. I have AES256 encryption on one partition and i cannot notice any delays even with very large files
19:51:38crwlSlasheri, what, you perhaps have illegal copy protection circumventing tools there?!
19:52:42Slashericrwl: Hehe :) I hate that new law..
19:53:12SlasheriI have ripped many cds to my iriver, but i am not sure if they were copyprotected (linux doesn't know if they are or not)
19:53:55_FireFly_if you make only private copys that shouldn't be the problem
19:54:39crwli don't have copy protected cd's, and i also try to avoid them
19:54:51crwlit's difficult to order cd's online because of that
19:55:26crwlsony has stated that they'll apply that second generation copy protection bullshit to all new albums somewhere next year
19:55:53Slasheri:/
19:56:39Slasherii don't think that new law will prevent piratism but instead it would encourage people to do illegal copying
19:56:41crwlthe new law is, sadly, a gross joke
19:56:47crwlalbeit a joke nobody understands completely
19:56:47Slasheriyep..
19:57:25***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:57:38***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:57:44***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:58:12***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:58:21***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:58:21***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:58:42***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
19:58:42***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( http://freenode.net/faq.shtml#privmsg )'
20:00
20:01:46_FireFly_ /msg nickserv set unfiltered on
20:03:42*amiconn summons Bagder
20:04:09 Join linuxstb [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net)
20:05:44 Join tephans [0] (n=stephan@p54A4473E.dip.t-dialin.net)
20:06:20 Quit tephans (Remote closed the connection)
20:20:20 Join muesli- [0] (i=muesli_t@Bbc99.b.pppool.de)
20:21:48muesli-re
20:21:58preglowi never did understand what 're' means
20:22:36 Join Bger_cgiirc [0] (n=54f2a148@labb.contactor.se)
20:22:58 Join DangerousDan [0] (n=Miranda@newtpulsifer.campus.luth.se)
20:23:11muesli-lol preglow..looks if this question turns into a running gag ;)
20:23:20muesli-should write a faq about it ;)
20:24:39linuxstbWhat does 're' mean?
20:24:53 Quit Bger_cgiirc (Client Quit)
20:25:12muesli-it stands for being back
20:25:22preglowwell, yeah, but why? :)
20:25:28muesli-dunno ;)
20:25:49muesli-its very common in german chats
20:25:59amiconnre(turned) ?
20:26:14muesli-at least this would make sense
20:26:36 Join Bger_ [0] (n=Bger_@free-dialup-84.242.161.72.Sofia.dialup.0rbitel.net)
20:27:00Bger_Slasheri: i don't think i've done something 'special' to slow it down
20:27:08linuxstbI'm in the process of making the endian functions as follows (please shout if you object):
20:27:48linuxstb1) Rename SWAB16 and SWAB32 to swap16 and swap32 - they simply do the byte-swapping when called.
20:28:39 Quit pabs (Remote closed the connection)
20:28:41 Join pabs [0] (n=pabs@xor.pablotron.org)
20:29:10linuxstb2) Implement the set of macros of the form letoh16 - for each type (big/little endian) of target, half of these are just identity macros.
20:29:12Bger_Slasheri: see http://www.cr0.net:8040/code/crypto/aes/aes.c at the end (after #ifdef TEST)
20:29:55linuxstb3) Replace the existing calls to SWAB16 and SWAB32 with the letoh32 (or the equivalent htole32) depending on the context used.
20:30:05Bger_linuxstb: for me it's perfect, but i'm not the one to say it
20:30:15amiconnlinuxstb: Sounds good. Needs general C implementation of swap16() and swap32()
20:30:18Bger_btw i think this is the right way
20:30:20linuxstb4) In ata.c, I'm just planning to change the calls to SWAB16 and SWAB32 to swap16 and swap32
20:30:27amiconn...used when no asm is avaliable (i.e. simulator)
20:30:40amiconn4) is WRONG
20:30:50linuxstbThere are already C implementations in system.h - I've copied them so they are compiled for the sim.
20:31:03Bger_linuxstb u must use your letoh32 etc
20:31:12linuxstbBger: Where?
20:31:17Bger_in ata.c
20:31:20amiconnSWAB16 and SWAB32 in ata.c MUST be replaced with the proper letoh/ htole macros
20:31:48linuxstbNo - they are already wrapped round SWAP_WORDS defines
20:32:08amiconnYes, but that's an entire different thing
20:32:26amiconnSWAP_WORDS tells whether the platform has byte-swap hardware for ata
20:32:38amiconniriver has, archos doesn't
20:32:44amiconnBoth are big endian
20:33:00linuxstbOK. If you're sure, then I'll do it.
20:33:09 Join webguest57 [0] (n=d4406110@labb.contactor.se)
20:33:21amiconnThe thing is that ATA is always little endian, because of its origin
20:33:51webguest57is there anyway thart I can add songs to a existing playlist ?
20:33:58webguest57thart = there
20:34:31Bger_amiconn: possible reasons for "I03: AddrErr" ?
20:34:45Bger_on iriver
20:34:47linuxstbAlso, there are places in ata.c that don't use the SWAB functions - it's not obvious why that is.
20:34:47webguest57er, a saved playlist not in play that is
20:35:07amiconnlinuxstb: Whic ones?
20:35:09amiconn+h
20:35:44amiconnBger: All sorts of reasons may cause this. I'd suspect a stack overflow if this is pure C code
20:36:11Bger_pure C
20:36:19Bger_very probable reason
20:36:22 Quit Birdfish (Read error: 104 (Connection reset by peer))
20:36:41Bger_btw, iriver variant with the 2 algorithms only is about 35k (iirc)
20:37:16Bger_amiconn thank u very much, gotta go
20:37:36linuxstbamiconn: around lines 349 and 663 in ata.c
20:37:58Bger_i'll look at this tomorrow ... nite all
20:38:45amiconnlinuxstb: Ah, these. These don't use SWAB16 because they deal with odd data alignment in memory
20:39:06linuxstbOK. This is why I didn't want to touch ata.c :)
20:39:13linuxstbBut it's a good way to understand it.
20:39:18amiconnThe SH1 doesn't handle this, it would cause an AddrErr exception
20:39:19 Join matsl [0] (n=matsl@1-1-4-2a.mal.sth.bostream.se)
20:39:31webguest57I would really like to add songs to a saved (not playing) playlist, is this possible ?
20:39:38amiconnThe coldfire does handle it, but probably slower
20:39:46amiconn...than doing the split beforehand
20:39:56 Quit Bger_ ()
20:40:17amiconnlinuxstb: ata.c is fun...
20:40:45amiconn...sometimes bugs are only found after years. Ask Linus ;)
20:41:25*webguest57 goes to make popcorn
20:41:28linuxstbAnyone fancy writing the iPod parts of it?
20:42:22amiconnI think this can only be done by an ipod owner among the rockbox hackers. Do you know one? ;)
20:43:21linuxstbAnyone want to buy an ipod?
20:45:20webguest57I'll buy your pod for info on the playlist thing
20:45:54amiconnI don't think adapting ata.c to ipod will be hard. The ipodlinux guys already did a lot, and their results are available as source code
20:46:22amiconn...unlike with the MMC driver for Ondio. I had to write that from scratch
20:46:45linuxstbYes - I'm not that worried about it. The complication is that their changes are just patches to the standard Linux IDE driver - so I need to dig into the Linux kernel to find out how it works.
20:47:32amiconnI agree that documenting the hardware would have been more helpful to others
20:47:47amiconnNot everyone is up to digging in linux kernel sources
20:48:57linuxstbAny idea how risky ATA driver development is - could I damage my ipod and/or disk if I do the wrong thing?
20:50:11amiconnI think 2 bad things can happen. (1) You can destroy all data on the disk. Better back up before, then that's no problem
20:50:50amiconn(2) You can accidentally lock your disk. Dunno how hard it would be to unlock it again, that depends on the disk model and other things
20:51:19linuxstb1) Isn't a problem - I obviously won't keep anything important on there.
20:51:23amiconnThis actually happened in very early stages of rockbox for archos, but I wasn't around yet by then
20:51:38linuxstbYes, I remember 2). Luckily I was never affected.
20:52:37linuxstbI wonder what Apple will say if I send them my ipod with a locked disk.
20:52:40amiconn(1) actually happened to me during MMC driver development, but not due to bugs in the MMC driver
20:53:27amiconnThe rockbox config sector was hardcoded to sector 61 then, but MMC uses a 32 sectors/track layout. The config sector ended up in the FAT....
20:54:41amiconnIt's always a good idea to start working on disk reading, and only advance to writing if reading is already working well
20:55:56 Join Sting [0] (n=Miranda@Toronto-HSE-ppp3878320.sympatico.ca)
20:56:31Stingwhee-ha! =) my first time on iRC
20:56:36 Part webguest57
20:57:26 Quit Sting (Client Quit)
21:00
21:01:24amiconnlinuxstb: Is arm big endian? I think it is...
21:01:26muesli-like the first time..very short fun ;)
21:01:42linuxstbamiconn: It's both.
21:02:02amiconnHuh? In what way?
21:02:09amiconnRun-time selectable?
21:02:11linuxstbWell, I mean it's either.
21:02:32linuxstbYes - run-time selectable. But I think big-endian is less efficient. So little-endian is the normal mode.
21:02:53linuxstbApple's firmware runs it in little-endian mode.
21:02:54amiconnHmm. Then ipod will be the first little endian target
21:03:12amiconnBug fixing fun...
21:03:24linuxstbYes, but I'm not worried about that. The Sims have always meant that at least the higher-level code has been well-written.
21:03:32amiconnYes
21:03:41linuxstbAnd I'm rewriting most of the low-level stuff anyway.
21:03:59linuxstbe.g. the LCD driver is 16-bit colour which is brand new for Rockbox.
21:04:12amiconnyes, but already prepared a bit :)
21:04:40linuxstbMmm. libmusepack defines its own swap32 function...
21:05:03preglowlibmusepack....
21:05:32amiconnlinuxstb: Speaking about codecs... any news concerning the alternative flac codec
21:05:33amiconn?
21:06:01linuxstbI started work on it last night.
21:06:33linuxstbI'm in the process of ripping it out of ffmpeg and writing a simple standalone decoder which I can run on my PC. I can then test how compatible it is.
21:06:44amiconnlibmusepack's config_types.h is 'fun'
21:06:58amiconnHow do they make sure 'int' is 32 bit??
21:07:09amiconnThey didn't know about calmrisc16...
21:07:50amiconn(flac) Sounds good...
21:10:26preglowlinuxstb: why not just try files out with vlc? that uses ffmpeg codecs
21:10:34amiconnHaha, the libmusepack swap32 macro looks rather inefficient to me
21:10:45linuxstbamiconn: Don't worry, I've deleted it.
21:10:58linuxstbMy contribution to libmusepack optimisation :)
21:11:18 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
21:11:25*amiconn should really start working on coldfire asm optimised memcpy()...
21:11:32preglowyes, he shoulkd
21:11:46linuxstbpreglow: You're right. I've already tested it on a few files using mplayer, and it is fine. So I'm probably past the testing stage now.
21:11:47amiconnBut first some plugin work
21:12:22amiconnThere are some in the tracker...
21:13:04preglowplasma :_P
21:13:17 Nick TiMiD[zzz] is now known as TiMiD (n=TiMiD[FD@asgard.valombre.net)
21:13:17DBUGEnqueued KICK TiMiD
21:13:20amiconnYes, and fire
21:13:26TiMiDhi
21:13:30amiconnAh TiMiD awoke ;)
21:13:31preglowfire isn't the best i've ever seen, but
21:13:49TiMiD;)
21:13:58amiconnI'll try both just now, on h1x0 and archos
21:14:06TiMiDTiMiD just recovered his screen
21:14:29CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:14:29*preglow Goto80 - Papaya Coconut
21:14:53TiMiDamiconn: you wanted to say me smth ?
21:16:00amiconnfire.rock is doing something strange...
21:16:10TiMiDhmm
21:16:13preglowlike how?
21:16:31amiconnThe bottom 4 pixel rows are looking odd (like pixel garbage) and it is slooo-oooow
21:16:37preglowon archos?
21:16:43TiMiDyes it's sloooooooooooooww
21:16:46TiMiD:)
21:16:47amiconnNo, H140 atm
21:16:53preglowamiconn: fire works like that, the bottom row is noise
21:16:57***Saving seen data "./dancer.seen"
21:17:02amiconnI have a suspicion why it is slow...
21:17:03preglowamiconn: and the rows above that are filtered versions of the row below it
21:17:06TiMiDI can remove the bottom lines
21:17:08linuxstbQuick CVS question - can I commit changes in both apps/ and firmware/ in the same commit?
21:17:17TiMiDamiconn: ah !
21:17:33amiconnlinuxstb: Yes of course
21:17:46TiMiDI always wanted to know how to do an optimized fire effect :)
21:17:46preglowahh, height of flame, why didn't i try that
21:18:27TiMiD(btw, the code is a little buggy, I have a new version that works slightly faster and with less bugsà)
21:18:31preglowbut now that we've started porint lots of oldschool demo effects to rockbox, where are the rotozoomers and toroids? :P
21:18:45preglows/porint/porting/(
21:19:26TiMiDpreglow: I was doing a 3d starfield simulator with rotations around x, y and z before I started working on remote
21:19:42preglowport second reality to rockbox!
21:19:44TiMiDhttp://timidzone.free.fr/pub/rockbox/starfield.rock
21:19:48TiMiDmaybe this one
21:21:27TiMiDI didn't ported the demos I did, I coded them from scratch (that's probably why they are so slo) (of course, for fire effect I didn't invented the princilples of the effect)
21:22:33preglowtime for a beer and a movie
21:23:05TiMiDhave good drinking / watching then :)
21:24:11TiMiDamiconn: so what was your idea to make fire run faster ?
21:27:38 Join Paul_The_Nerd [0] (n=paulthen@cpe-66-68-93-2.austin.res.rr.com)
21:27:57linuxstbOK, byte-swap functions now committed.
21:30:12Paul_The_NerdGood day everyone.
21:30:41TiMiDwelcome !
21:32:03_FireFly_hiho :)
21:33:35 Quit ghode|afk (Read error: 110 (Connection timed out))
21:33:53 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
21:35:29_FireFly_I think the remote is mostly done, only the playlist-viewer in addition to the debug-menu-entrys and the plugins aren't displaying something on the remote
21:35:56linuxstbHas anyone tried compiling the plasma plugin with the current CVS Rockbox? It doesn't work on my H140 - it freezes on the Please Wait... splash.
21:36:07TiMiDuh ?
21:36:15TiMiDthere is a please wait oO
21:36:24TiMiDI thought I removed it ....
21:36:33linuxstbMaybe I'm using an old version then.
21:36:34Paul_The_Nerd_FireFly_: Could you kick the remote back to the WPS (if music is playing) when a plugin loads? As a temporary solution, at least?
21:36:36_FireFly_I think for the debug-menu there no need to display it on the remote also for most of the plugins
21:36:47linuxstbTiMiD: Should I just remove it from the source then?
21:36:59TiMiDno
21:36:59TiMiDwait
21:37:05TiMiDI will upload the newer version
21:37:30_FireFly_Paul_The_Nerd: I don't know if this is possible with the current code
21:37:34linuxstbThanks. Preglow is pressuring for it to be comitted :)
21:37:36TiMiD(You must have old sources
21:37:38Paul_The_NerdAaah
21:37:56linuxstbTiMiD: I downloaded it from the patch tracker a day or two ago.
21:39:13TiMiDmaybe I updated it at home and didn't uploaded it :)
21:39:47linuxstbThe version I have downloaded is dated 2005-08-31 and is 6830 bytes in size.
21:40:35TiMiDhttp://timidzone.free.fr/pub/rockbox/
21:40:38TiMiDlook here
21:40:45linuxstbI must have an old version - I've just re-downloaded it from the patch tracker and it is different.
21:40:56TiMiD(the .rocks ar not up to date)
21:41:08TiMiDlast version should be here
21:41:23TiMiD(not on the patch tracker XD
21:42:36linuxstbOK, got it now. I had a very old version in my plugins directory - my mistake. Sorry.
21:45:07TiMiDthere is a little thing to do to the plugin : test it on archos
21:45:36TiMiD(and maybe remap the keys)
21:45:44TiMiDshould be the same for fire
21:45:52Paul_The_NerdI have an outdated plasma.rock too, though mine's interesting.
21:45:52amiconnPlasma is working on archos, but fire is not
21:45:58TiMiD(however, I wonder how slow will fire run on archos XD)
21:45:59Paul_The_NerdIt displays in monochrome... or maybe 2-bit grayscale
21:46:02Paul_The_NerdI can't tell for sure.
21:46:14amiconn...and I'm looking into the speed issue atm
21:46:33linuxstbamiconn: Are you preparing to commit plasma?
21:47:15TiMiDamiconn: excepted speed, what is not working on archos with fire ?
21:47:25amiconnIt says 'Not enough memory'
21:47:26TiMiD(speed and compilation problems :p)
21:47:30TiMiDouch :)
21:47:39TiMiDIt's strange
21:48:01TiMiDif I remember well, it should just use as much memory as plasma
21:48:13amiconnYes, it's obvious that you can't reserve a 32-plane grayscale buffer when the plugin already takes 25KB out of the available 32KB
21:48:16TiMiD(the only big thing in memory is the screen buffer)
21:48:53amiconn...and it seems there is a way to considerably speed up plasma
21:49:03TiMiD:):):):):)
21:49:21TiMiDI can't wait :p
21:49:21amiconn...by using the buffered mode of the grayscale lib
21:49:52amiconnThe pixels are calculated individually anyway, and buffer mode only touches pixels which really have changed
21:50:09amiconn..but it takes more ram for the buffers
21:50:19amiconnMaybe it won't work on archos because of this...
21:50:25_FireFly_Want someone make the playlist-viewer displaying to the remote ??
21:50:43TiMiDfor fire, 9when I tested, 90% of cpu time was used to compute the effect
21:51:10 Quit muesli- (Read error: 110 (Connection timed out))
21:51:13TiMiD_FireFly_: you coded it ?
21:51:25amiconnPerhaps fire should boost the cpu, because the grayscale lib does no longer do it automatically
21:51:39TiMiDeven with cpu boosted it's slow
21:51:41_FireFly_TiMiD: ??
21:52:16TiMiDand even on my amd 2ghz with SDL its slow (when reaching high resolutions though)
21:52:46TiMiD_FireFly_: if you havn't already coded it, it would be great to do it when You know if my patch is accepted or not
21:53:04 Quit Maxime (Read error: 104 (Connection reset by peer))
21:53:04TiMiDin order not to make the work twice ;)
21:53:47 Join Maxime [0] (n=flemmard@fbx.flemmard.net)
21:54:18_FireFly_the playlist-viewer is the last thing in additon to the debug-menu-entrys and the plugins which doesn't display something on the remote :)
21:54:29_FireFly_the rest is working
21:54:53_FireFly_TiMiD: how far are you with your widget system ??
21:55:39TiMiDone last thing to port to be able to commit it without any functionnality lost
21:55:46TiMiDbutton abr on archos
21:55:52TiMiDbutton bar
22:00
22:01:24 Join linuxstb_ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net)
22:03:10_FireFly_TiMiD: In your build on your webspace the menu isn't displayed on the remote
22:04:17TiMiDyes
22:04:27TiMiDI only modified the file viewer
22:04:59TiMiDbut the menu system is a lot less compicated so it will be a piece of cake :)
22:05:30TiMiD(but I want to know first if this is going to get commited or not
22:05:40TiMiDto avoid waisting my time :p
22:06:57_FireFly_i had just an unreproduceable bug
22:07:27_FireFly_after leaving the wps the file-tree was displaying only the line-selector
22:07:55TiMiDso bad :(
22:08:09TiMiDwas it on boths screens ?
22:08:33TiMiD(and whith which build did it happened ?)
22:08:44_FireFly_I'm sorry i had only lookd on the main-lcd
22:08:50TiMiDok
22:09:58_FireFly_it displays 051006-0202 as version
22:10:01TiMiD(if it wasn't today, it may be an already soved issue because I solved some bugs yesterday and put a new version with pgup / pgdow support on my ftp)
22:10:18TiMiDso it's form today's build :(
22:10:21_FireFly_it was today i have just tested your build
22:11:05TiMiDso, when you were moving the cursor, it just displayed what was selected and the rest was blank ?
22:11:53_FireFly_complete blank only the tiny triangle-line-selector was shown
22:12:17TiMiDand it was moving right?
22:12:24_FireFly_yepp
22:12:29_FireFly_it was move-able
22:12:39TiMiDwer statusbar and scroll bar still alive ?
22:13:23_FireFly_the tree itself seemed to there but not shown the statusbar was on scrollbar i don't know because it wasn't necessary to display an scrollbar
22:13:29_FireFly_to be there
22:14:06TiMiDok
22:14:17TiMiDand you could still select items and play them ?
22:15:06TiMiDwell I'm clueless
22:15:28_FireFly_i'm too because i can't reproduce it
22:15:37 Join linuxstb__ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net)
22:15:38TiMiDbecause the only moment when you can see the cursor is when all the text lines before it have been displayed :)
22:16:39TiMiD(just look at the drawing fn in gui_list.c if you are courageous enough :p
22:17:25TiMiDit's also possible that nothing was displayed because it receives nothing to display (in that case, it's more compilcated ^^)
22:18:41 Quit linuxstb (Read error: 110 (Connection timed out))
22:23:56_FireFly_i hope that was only a inconsistens in the config
22:26:11TiMiDI will keep an eye on it :)
22:26:40_FireFly_ah maybe i have the reason why it happend
22:26:42TiMiDand it have better not to do this to me, if not ... I will punish him ^^
22:27:51TiMiDyes tell me more :)
22:27:51_FireFly_i have just installed my version and after starting i got a blank file-tree where only the line-selector was drawn and some part of the statusbar
22:28:07_FireFly_but tree wasn't working
22:28:11 Quit linuxstb_ (Read error: 110 (Connection timed out))
22:28:11 Join lImbus [0] (i=lImbus@port-212-202-8-79.dynamic.qsc.de)
22:28:24TiMiDso it's also with the current rb version :p
22:28:28lImbushi all
22:28:35TiMiDhi lImbus
22:28:39_FireFly_after loading a config file the problem was gone ;=)
22:29:15TiMiDpffiewww :)
22:29:16_FireFly_now testing your build again
22:29:27TiMiDmaybe one bug less :p
22:30:12amiconnTiMiD: fire.rock works on archos (without the 32-planes check), but slow, and only using a few shades
22:30:19amiconn(guess 7 or 8 shades)
22:30:38_FireFly_TiMiD: ok it's a config inconsistence
22:30:48amiconnIt's not as slow as I would expect though, a bit more than one refresh per second
22:31:31amiconnA big performance eater is the array double-subscription
22:31:50TiMiD2 dimension array ?
22:31:54amiconnThe gcc generated asm calculates this for every loop iteration!
22:32:09amiconnI mean using array[x][y] within the loop
22:32:22amiconnYOu're almost always better off using pointers
22:32:55amiconnOh, and btw, tab_init_0() can be replaced with a simple rb->memset() call
22:33:00TiMiDyes, I knew it was those line which was eating all the cpu (because I tried to remove all array-related code and it ran 10x faster :p)
22:33:16TiMiDbut I don't have a clue on how to do without it
22:33:27amiconnI will try
22:33:36TiMiDok thx :)
22:33:37_FireFly_TiMiD: you have reset the config or load a config-file and then you must reboot the device
22:33:51TiMiD(and try not to put too much assembly in it ;) )
22:34:00amiconnAnd for the bottom 4 rows - we can just make the fire[] array 4 pixels higher than the visible area
22:34:20TiMiDyes, it will be easy in the fn which applies the palete :p
22:34:34_FireFly_to reset
22:35:07TiMiD_FireFly_: does it do this with unmodified builds ?
22:35:24_FireFly_i will test
22:35:29TiMiDok thx :p
22:36:21amiconnTiMiD: I'll flip all arrays to row-wise addressing, i.e. array[x][y] will become array[y][x]
22:36:57TiMiDok
22:37:15TiMiDI'm very curious to see how you 'll manage to do this :)
22:39:01TiMiD(btw, if the grayscale were uniformly distributed between 0 and 255, it would avoid a palette applying loop :) )
22:39:13TiMiDbut that's only on iriver
22:39:26amiconnThey are uniformly distributed, but the lcd isn't linear
22:40:08_FireFly_ok the problem doesn't appear when using an unmodified daily build
22:41:24_FireFly_then my added options in my remote-patch is the trigger for the "problem"
22:41:38_FireFly_s/then/so
22:41:51TiMiD_FireFly_: so your build does the same :)
22:42:02TiMiDwe are both in the same boat ^^
22:42:18_FireFly_no you misunderstood me
22:42:31TiMiDamiconn: would it be possibleto correct that iriver problem in the grayscal lib itself ?
22:43:12TiMiD_FireFly_: sry, I'm not native speaker and I misundertand often ;p
22:43:13_FireFly_when i "update" a unmodified daily-build with your build then the problem doesn't occurs
22:43:20TiMiDok
22:44:03_FireFly_i have added some options for the remote(e.g.sperate status-/scrollbar) in the global_settings struct
22:44:07_FireFly_in my patch
22:44:19TiMiDok Iunderstand :)
22:44:28_FireFly_:)
22:44:59TiMiDand what if you put your config files from your modified build to an unmodified one ?
22:46:51_FireFly_it's not the config-files it's the config which is saved in the RTC-Ram on shutdown
22:47:26 Join ghode|afk [0] (n=garudin@host-83-146-62-103.bulldogdsl.com)
22:48:50TiMiDoh
22:49:07TiMiDI didn't knew it was saved that way
22:49:20TiMiDI always thought it was in a config file
22:49:50_FireFly_search for config_block in the sources :)
22:50:41amiconnThere is no rtc ram on h1x0
22:51:00_FireFly_hmm where will be then the config saved ??
22:51:28amiconnThe active configuration is saved to the config sector on all units, and on units with rtc _a part_ of the settings is saved to the rtc ram
22:51:51_FireFly_oh then i mean the config-sector
22:52:00TiMiDon the disk ?
22:52:04amiconnyes
22:52:10_FireFly_but it is no file
22:52:14TiMiDwhy not in a file ?
22:52:22TiMiDit would be more convenient
22:52:35_FireFly_i think it is a part of the track 0 of the disk
22:52:35amiconnIt's way simpler this way
22:52:51amiconnThe config sector is start_of_first_partition - 2, if that exists and is unoccupied
22:53:13amiconnPraktically always sector 61 on track 0
22:53:23amiconn(on harddisk units)
22:53:29TiMiDyes
22:53:34amiconnOn Ondio it is usually sector 30
22:53:46TiMiDbut a file only require a file access
22:53:51XavierGryes but if the config was on a file we could have theoritically infinite nymber of settings.
22:54:00_FireFly_and so the problem can occure when i switch from my build to the clean daily-build
22:54:18TiMiD_FireFly_: glad to hear it :)
22:54:57_FireFly_because there is no refer which byte(s) is which option
22:54:58amiconnTiMiD: File access has way more overhead than sector access. Files should be human readable, so reading and parsing ascii is necessary. We can run out of diskspace while writing files etc etc
22:55:29amiconn_FireFly_: That's what the config block version is for. When it doesn't match, the settings are reset to default
22:55:55_FireFly_hmm ok then i have forgotten to change the version
22:55:55XavierGralso amiconn why did the devs choose Rockbox to remember settings files only in the .rockbox folders? (Font, config, lang, wps )
22:55:56amiconnRemember numerous commit messages saying 'Bumped config block version, so save your settings'?
22:56:09XavierGrIs that the sector will be overrun with data?
22:56:25_FireFly_this messages have i badly ignored
22:56:26amiconnXavierGr: That's obvious: it avoids storing the path, hence saves space
22:56:51XavierGryes I know I just did the same with fmr files, just qurious
22:57:06XavierGr^curious
22:57:09TiMiDI'm not convinced :)
22:57:11amiconnI'd actually consider that a feature
22:57:27TiMiDbut if it works ...
22:57:30linuxstb__Am I misunderstanding, or does line 357 of firmware/pcm_record.h swap the left and right channels? (both before and after my last commit).
22:57:52XavierGrsomeone has already mentioned this, I think
22:58:19linuxstb__i.e. I think it was assuming that SWAB32 was complying with the Posix definition.
23:00
23:01:07amiconnThere is a posix definition of this?
23:01:21amiconnI think you're right, it does swap channels
23:01:32linuxstb__Well, "swab" means "swap adjacent bytes". So applying it to 4 bytes will do what the author of pcm_record wanted.
23:01:47amiconnIirc some people already complained that iriver recording swaps channels...
23:02:06linuxstb__That seems confirmed then. I'll fix it.
23:02:12amiconnSome remarks concerning your changes:
23:02:29linuxstb__Happy to hear them.
23:03:01amiconnIn fat.c, you sometimes confused letohXX and htoleXX. Doesn't hurt, but isn't logical
23:03:19linuxstb__That's because I didn't fully understand the code. So I just guessed in some places.
23:03:30linuxstb__(knowing that the effect is identical)
23:03:43linuxstb__Please fix them (or tell me where to fix them).
23:04:08amiconn(2) Did you remove SWAW32()? It's inconsistent with the rest now, and not used anywhere (iirc)
23:04:31linuxstb__I noticed that function, and also the fact that it isn't used anywhere. So I decided to leave it.
23:04:44linuxstb__But we could delete it.
23:05:40amiconnyes
23:06:37amiconnDid you check musepack speed?
23:06:54linuxstb__No, I don't have any files available.
23:07:26amiconnHmm, me neither. Could be interesting...
23:07:32linuxstb__I didn't look - was that swap32 function used a lot in musepack?
23:07:51amiconnbut maybe swap32 isn't used in the actual decoder
23:08:17amiconnHmm, ata.c is indeed a bit ambiguous
23:08:53amiconnThe problem is that the iriver, while being big endian, handles ata like a little endian machine because it has a hardware byte swap
23:09:29amiconn...and the identify data is *the* weird exception from the rule that ATA is little endian
23:10:58XavierGrwhy dont they choose one endianess to be sure, even if there is a slight speed boost for some tasks, is it so big to use 2 endianess's?
23:11:30linuxstb__Who are "they" ?
23:11:33XavierGrI generally (like metrics)
23:11:37XavierGr^mean
23:11:57XavierGrthey I mean all companies that make hardware.
23:12:11XavierGr(though I am ignorant on this)
23:12:47linuxstb__I've no idea. I can understand one generation of chips being the same endianness as the previous generation. But those choices were made in the 1970s.
23:13:40XavierGrI know that for some tasks big-endian is better and for other little will do more but is it worth to have each time to check and implement swap functions that may cause hidden errors?
23:13:49amiconnlinuxstb__: fat.c, lines 822 and 826 should be letoh16()
23:14:23amiconnDito lines 857 and 862 should be letoh32()
23:14:29XavierGrIt's like screws. They agreed which way will loose it.
23:15:30linuxstb__amiconn: Thanks.
23:15:47_FireFly_night everybody
23:15:48 Quit ender` (Read error: 104 (Connection reset by peer))
23:15:55XavierGrnight!
23:16:00 Quit _FireFly_ ("Leaving")
23:16:16amiconnXavierGr: In fact no endianess is better than the other. It's just because there are two, some tasks are slower because of the necessary conversion
23:16:59***Saving seen data "./dancer.seen"
23:17:05XavierGrI didn't said that on is better, I have heard that a specific endianess may be better for specific tasks.
23:17:25amiconnYes, but that is due to the data the task uses
23:17:38XavierGryes
23:18:04amiconnE.g. network data is always big endian, so it's more efficient to process it on a big endian CPU
23:18:25XavierGrso my question remains why have 2? Is it necessary
23:18:55amiconnIt's not
23:18:57XavierGryes but will it have speed problems on a little endian.
23:19:20XavierGrI mean ok it will run a tiny slower so what.
23:19:23amiconn...but they are there, and we have to deal with them
23:19:38XavierGryeah I just comment it.
23:20:10Paul_The_NerdBut why would you use multiple endians withing one device?
23:20:18Paul_The_Nerdwithin
23:20:25amiconnIf only one endianess would be in use, all data would use the appropriate format
23:20:47amiconnPaul_The_Nerd: Sometimes you have to, e.g. all ATA hardware is little endian
23:20:59XavierGrLike metrics, americans have different metrics from the europeans, and when NASA collaborated with a europian company they crashed their satelite because the american program was programmed to take different input on metrics.
23:21:01amiconn...but a number of embedded CPUs is big endian
23:23:06 Join _aLF [0] (n=Alexandr@mut38-2-82-67-66-128.fbx.proxad.net)
23:24:27linuxstb__pcm_record.c should be fixed (no more swapping of channels) and I've corrected fat.c
23:24:39amiconngoodie
23:27:19XavierGrcould someone delete patch 1312132? I was an idiot when submitting and forgot to log on. Now I want to update the patch and I can't.
23:27:39XavierGrSo in order to avoid duplicates can someone delete it?
23:27:53 Join webguest14 [0] (n=d4406110@labb.contactor.se)
23:28:43amiconnXavierGr: Is there a new one?
23:29:19XavierGrwell LinusN checked it this morning and I fixed some things.
23:29:25webguest14XavierGr:, not entirely true, it was a mixup between, imperial inches and metric that causec that probe to crash http://edition.cnn.com/TECH/space/9909/30/mars.metric.02/
23:30:18XavierGryes I didn't remembered that right, but again it is quite the same and hilarious.
23:31:07TiMiDso the britanic are guilty :)
23:31:29XavierGrmaybe yes, i didn;t accused the americans
23:31:32webguest14yep the dear brits that have only recently gone to metric
23:31:47XavierGrbut both off them are just stubborn with metrics and measurments
23:32:11webguest14the yanks have been 'oficialy' metric for years
23:32:39webguest14but its not enforced
23:35:25webguest14the average yank still uses farenheit for temperature, they cant grasp that water freezes at 0 and boils at 100
23:36:28 Part webguest14
23:36:44XavierGrok new patch submitted so you can delete the old.
23:40:41 Quit matsl (Remote closed the connection)
23:41:49 Quit Paul_The_Nerd ("Chatzilla 0.9.68a [Firefox 1.0.7/20050915]")
23:41:56amiconnTiMiD: I know why the top 4 rows are garbage. Check the big loop in fire_generate(): you're using data from lines y+1, y+2 and y+3
23:42:10amiconn...unconditionally, and up to y==LCD_HEIGHT-1
23:42:21amiconnThat means you're reading out-of-bounds...
23:43:45TiMiDamiconn: I know, this is an alod version
23:43:49TiMiDold version
23:44:08amiconnHmpf, why didn't you update the patch in the tracker?
23:44:09TiMiDIf you want I can give you the sdl version
23:44:17TiMiDwhich is more recent
23:44:33TiMiD(but under linux so I will have to reboo)
23:45:22TiMiDI didn't updated on the tracker ... because I thought nobody wanted to include it in cvs
23:45:34*amiconn thinks he's wasting his time now :/
23:45:36TiMiDand since it's not a major release ...
23:45:41TiMiDnope
23:45:56TiMiDthe new version isn't that far from this ome
23:45:57TiMiDone
23:48:53TiMiDoops I didn't fixed it in the SDL version too (I thought I did it ...)
23:48:57TiMiD pixel_value-=cooling_map[ x ][ y ];
23:48:57TiMiD if(pixel_value<0)
23:48:57TiMiD pixel_value=0;
23:48:57TiMiD else if(pixel_value>255)
23:48:57TiMiD pixel_value=255;
23:49:00TiMiD fire[x][y]=pixel_value;
23:49:06TiMiDthat's what changed
23:49:36TiMiDbut since with SDL, I took integers, I do'nt think it's appropriate on iriver
23:50:34TiMiD(well you must just avoid to averflow since there is no test, the value goes from 255->0 and it makes strange bugs sometimes)
23:51:12amiconnIt might actually be faster to use 'int' for the accumulation variable instead of char
23:51:26amiconnOn target, this often yields more compact binary code
23:51:27TiMiDjust make pixel_value an integer and test it before assigningit's value to the buffer
23:51:38TiMiDok
23:52:01TiMiDwell I think you 'll be alright with this :)
23:52:27XavierGramiconn can you remind me the button handling for Ondio on the JPEG file scrolling plugin?
23:52:28amiconnThat's what I am usually doing when something is slow: Compile, disassemble binary, spot the function, and compare the asm code with my C original
23:52:37XavierGrfor the mode i have set it to:
23:52:57TiMiDfor the out-of-bound y, a seaprated loop would be appropriate maybe becauste you don't want to use modulos right :)
23:52:59XavierGr#define JPEG_ZOOM_PRE BUTTON_MENU
23:52:59XavierGr#define JPEG_ZOOM_IN (BUTTON_MENU | BUTTON_REL)
23:52:59XavierGr#define JPEG_ZOOM_OUT (BUTTON_MENU | BUTTON_REPEAT)
23:52:59DBUGEnqueued KICK XavierGr
23:52:59XavierGr#define JPEG_NEXT (BUTTON_MENU | BUTTON_RIGHT)
23:53:06amiconnSometimes gcc is really clever, and sometimes it generates brain-dead code
23:53:48XavierGrbut I think I write it wrong
23:53:48TiMiDmode ?
23:54:01TiMiDyes JPEG is unrelated :)
23:54:34XavierGrehmm I think you told me ZOOM_IN = MENU + DOWN?
23:54:35TiMiDI was wondering what I didn't understood but this time it was not me :)
23:54:46XavierGrI can't remember
23:54:52TiMiDouch
23:54:58XavierGrlol
23:54:59TiMiDforget about all I said
23:55:15TiMiDi misread and thought XavierGr was amiconn
23:55:16amiconn#define JPEG_ZOOM_IN_PRE BUTTON_MENU
23:55:22amiconn#define JPEG_ZOOM_IN (BUTTON_MENU | BUTTON_REL)
23:55:33amiconn#define JPEG_ZOOM_OUT (BUTTON_MENU | BUTTON_DOWN)
23:55:48amiconn#define JPEG_NEXT (BUTTON_MENU | BUTTON_RIGHT)
23:55:57 Join webguest29 [0] (n=d49f4cf2@labb.contactor.se)
23:55:58amiconn#define JPEG_PREV (BUTTON_MENU | BUTTON_LEFT)
23:56:03XavierGryes thanks!
23:56:23XavierGrI missed the menu+down now I fixed it and made the code policing that Linus told me.
23:56:29amiconn...and you have to change the checks for JPEG_ZOOM_IN and JPEG_ZOOM_OUT
23:56:50XavierGrchecks?
23:57:37XavierGr#define JPEG_ZOOM_PRE BUTTON_MENU#define JPEG_ZOOM_IN (BUTTON_MENU | BUTTON_REL)#define JPEG_ZOOM_OUT (BUTTON_MENU | BUTTON_DOWN)#define JPEG_NEXT (BUTTON_MENU | BUTTON_RIGHT)#define JPEG_PREVIOUS (BUTTON_MENU | BUTTON_LEFT)
23:57:41XavierGrgrrrr
23:57:49XavierGrchange what and sorry for that

Previous day | Next day