--- Log for 03.08.108 Server: calvino.freenode.net Channel: #rockbox --- Nick: logbot Version: Dancer V4.16 Started: 1 day and 18 hours ago 00.00.54 Join fyre^OS [0] (n=fyre@cpe-68-173-172-243.nyc.res.rr.com) 00.02.34 # I still think the whole point of the procedure is to have two versions of Rockbox available, one in ROM and one in RAM. Since the patch isn't needed anymore, you don't need to have two versions around and just unziping a current rockbox.zip to the CF should work, as long as you have a current bootloader installed. 00.04.52 # I just don't understand the point where the wiki says to flash "rockbox.iriver (RAM image) to ROM". Opening the file with the flash plugin doesn't offer a choice where the flash it, I guess rockbox.iriver always ends up in RAM. 00.05.13 Quit fyrestorm (Read error: 60 (Operation timed out)) 00.06.23 # What's also strange is that when I choose "boot from disk" on the bootloader menu, I don't get "ata -80" but "no partitions found" instead. 00.06.29 Quit snoh (Remote closed the connection) 00.06.40 # But there is exactly one partition on the CF card. 00.09.19 Quit bluebrother ("leaving") 00.16.12 # rockbox.iriver always ends up in ram 00.16.22 # or rather, as the ram image 00.16.35 # the one called rombox is the one that is run direct from rom 00.17.11 # I can't remember the exact name, but it is something like rombox.iriver 00.18.52 # BigBambi: yes, it is. 00.19.24 # The thing is, both are stored in ROM 00.19.48 # The difference is that rockbox.iriver is copied to RAM to be run, rombox.iriver is run direct from ROM 00.20.24 # So when the wiki says to flash rockbox.iriver (RAM image) to ROM it is correct, if not too clear 00.20.30 Join kugel [0] (n=chatzill@unaffiliated/kugel) 00.22.21 # BigBambi: ah, so both are *stored* in ROM, but while rombox.iriver is directly run from ROM, rockbox.iriver is loaded into RAM before it is run? 00.22.27 # yes 00.23.39 # ah, thanks 00.27.39 # Anybody with experience running Rockbox from CF on Iriver H120 in here? 00.28.20 Quit Schmogel (Read error: 104 (Connection reset by peer)) 00.29.22 # http://www.rockbox.org/twiki/bin/rdiff/Main/WpsIpod5g?type=history causes an error 500. 00.29.46 # I know it is a large page with lots of revisions, but being unable to view history kinda sucks. 00.32.11 Join coatman [0] (n=Jacob@ppp-70-253-191-136.dsl.stlsmo.swbell.net) 00.34.18 Quit herrwaldo (Remote closed the connection) 00.35.09 Join herrwaldo [0] (n=waldo@ip-81-11-206-57.dsl.scarlet.be) 00.49.32 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul) 00.52.36 Quit bertrik ("Leaving") 00.54.46 Quit herrwaldo (Remote closed the connection) 00.55.17 Join herrwaldo [0] (n=waldo@ip-81-11-206-57.dsl.scarlet.be) 01.08.12 Join nevyn [0] (n=nevyn@210-84-52-241.dyn.iinet.net.au) 01.10.20 # Can I be reasonably confident that the following item is an ipod that is supported in rockbox? http://www.dealsdirect.com.au/p/60gb-apple-ipod-classic-black-reconfigured/ 01.10.27 *** Saving seen data "./dancer.seen" 01.12.45 # nevyn: ipod classic does not work with rockbox.. 01.12.59 # right. 01.13.07 # but aiui it's not actually an ipod classic 01.13.23 # Yea, that doesn't look like a classig 01.13.34 # this is part of why I'm asking 01.13.49 # aiui the classic is only the 80/160GB models 01.14.16 # nevyn: That sounds like refurbished ipod to me, where "reconfigured" means that the front and backplate have been replaced with original parts if they've taken damage 01.14.31 # kugel: that's what I think 01.14.34 # nevyn: That is true, but it might have had an odd hard drive stuck in 01.14.44 # in which case it's a g5.5 01.14.45 # However, I think it isn't a classic and will be OK 01.15.02 # nevyn: For reference, http://support.apple.com/kb/HT1353 01.15.35 # they list the model, http://www.google.com/search?hl=en&q=MA147LL 01.16.43 # Yes, googling for the model number brings up an ipod video 01.17.11 # neato 01.17.14 # So it should be fine, it isn't actually a classic 01.17.24 # BigBambi: that's what I figgured. 01.17.28 # No guarantees though :) 01.17.35 # of course. 01.17.53 # problem is. 01.18.04 # to the vast majority an ipod is an ipod is an ipod 01.18.12 # er, what? 01.18.23 # Not in hardware terms between generations 01.18.27 # sure. 01.18.39 # Sorry, I misread 01.18.40 # I know that. 01.19.07 # I missed the to at the start and inserted an of between majority and an 01.19.11 # but the labeling and so on don't make it clear which generation ipod things are. 01.19.53 # the vast majority of an ipod is an ipod is an ipod reads very differently :) 01.20.01 # heh 01.20.08 # nevyn: True, my friends even freak out if they see a ipod shuffle 01.20.09 # no to the vast majority of peopl 01.20.29 # I tell'em "but it has no display, that sucks", and they just answer: "but it's an ipod" 01.20.58 # and classic has become a way to specify the "standard" ipod as opposed to a touch or nano or shuffle 01.21.50 # thanks 01.22.38 # nevyn: But, the ipod video is actually the target on which rockbox performs worst 01.25.21 Join Zacam [0] (n=Zacam@fifth.penguinness.org) 01.26.13 Part domonoky 01.27.37 # kugel: bother. 01.27.45 # Big screen, slow processor 01.27.53 # what should I be looking for in this size storage range? 01.28.06 # www.rockbox.org/wiki/BuyersGuide 01.30.18 # ok so that doesn't mention that the ipod video sucks for rockbox. and it doesn't give a guide as to what works well/best 01.31.08 # portalplayer have fairly slow processors (ipod, h10, sansa), or the bigger the screen within those the worse 01.32.02 # It depends what you want to do - the ipod video will be likely fine unless you want to have every bit of dsp turned on (equalizer etc) or watch video 01.32.11 Quit BHSPitLappy (Read error: 110 (Connection timed out)) 01.32.52 # On the ipod video if you want to watch video you are better of using the apple firmware, as it can use an additional chip for video playback that we don't know how to use 01.33.57 # And I don't mean just equaliser, I mean all the other DSP options too (disclaimer, I've not actually used one, I'm just going on other people's experiences) 01.34.10 # The gigabeat F is very good though 01.35.56 Quit saratoga ("CGI:IRC (EOF)") 01.37.22 Join tonohono [0] (i=root@gingerale.dreamhost.com) 01.39.04 Quit advcomp2019 ("IRC is just multiplayer notepad") 01.39.42 Quit herrwaldo ("Konversation terminated!") 01.39.52 Join ackers [0] (n=Becks@80-235-149-109.cable.ubr22.newt.blueyonder.co.uk) 01.39.53 # has anyone actually connected a jtag debugger to the e200 successfully? 01.40.32 Quit ompaul (Client Quit) 01.42.07 Part ackers 01.51.20 Quit ender` (" “That’s right, babe,” Cochrane chortled. “We’re not possessors, we’re just like dimensionally disadvantaged.” -- Peter F. H") 01.51.59 Join advcomp2019 [0] (n=advcomp2@66.172.233.113) 01.57.20 Part tonohono ("★") 01.58.39 Quit perrikwp ("http://www.mibbit.com ajax IRC Client") 02.06.25 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/mibbit.com/x-e17f72bddbe9c82a) 02.06.52 Part raphi 02.07.23 Join ChristopherW [0] (n=christop@ip68-3-205-51.ph.ph.cox.net) 02.08.11 Join thomc [0] (n=thomc@host86-159-61-118.range86-159.btcentralplus.com) 02.11.40 Part thomc 02.12.01 Join hd [0] (n=jd@modemcable234.136-203-24.mc.videotron.ca) 02.13.42 Quit HellDragon (Read error: 113 (No route to host)) 02.18.09 Nick hd is now known as HellDragon (n=jd@Wikipedia/HellDragon) 02.23.53 Join libre [0] (n=libre@93-97-13-142.zone5.bethere.co.uk) 02.24.29 Quit coatman (Read error: 104 (Connection reset by peer)) 02.26.27 Nick num1_away is now known as num1 (n=brian@c-68-63-193-45.hsd1.az.comcast.net) 02.29.33 Part libre ("Leaving") 02.48.21 Join Colby_ [0] (n=koru@c-67-169-234-72.hsd1.ca.comcast.net) 02.50.04 Join sitwon [0] (n=adam@pool-71-178-171-238.washdc.fios.verizon.net) 02.52.27 # Question: I have an iriver H10 20gb. Ages ago it got wet or something, and the scroll/track bar on it now doesn't scroll at all with iRiver firmware, and with Rockbox firmware it scrolls straight down no matter what (though it will scroll up, but then promptly back down). Also, whenever I try to use -any- buttons on my player with Rockbox (power, skip buttons, select, back) they don't work... 02.52.29 # ...at all - to the point where I have to manually reset my player to get it to even turn off. Any ideas? Is this a software or hardware issue? 02.54.23 # if they dont work properly in either firmware, it is hardware... 02.55.39 # I don't understand why I can't even browse the folder trees in Rockbox though - that at least works in the iRiver firmware. 02.56.06 # For me, the Rockbox menu won't even do anything except let me scroll between Files - System. I can't open any of those options. 03.04.45 Quit Colby_ ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]") 03.05.28 # IIRC Rockbox's scroll pad "interpreter" is significantly different than iRiver's, still it is clearly a hardware issue. 03.06.54 Quit Bensawsome ("The awsome is gone :(") 03.07.43 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome) 03.10.28 *** Saving seen data "./dancer.seen" 03.12.35 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon) 03.17.35 # someone mentioned that you can dynamically allocate memory on rockbox without using malloc 03.18.04 # How do I do this? (sorry for using multiple lines, my finger slipped) 03.19.49 # depends how much you want, and how dynamic, and how you want to use it 03.22.56 Join Thundercloud [0] (n=thunderc@84-51-130-71.judith186.adsl.metronet.co.uk) 03.26.49 # I want enough to hold an image the size of the screen, at the screens bitdepth. I basically want to make an array with a size determined at runtime. preferably read and write, unless you have a different meaning for "how you want to use it" 03.28.48 # on a stylistic note, in apps/lang there's a mix of naming a language file based on the word for the language, and the languages word for the language. If I make a patch that changes all the names to english what are the chances of it being accepted? 03.29.12 # any reason to not just make it statically sized at compile time? 03.29.32 # and there has been a bit of discussoin about that and we dont really knwow hat we want 03.30.50 # JdGordon: Morning :) 03.31.14 # hi 03.31.40 # I've looked... depending on my day ill concider commiting :p 03.31.49 # I would prefer to save space, but that's not really a strong argument at this scale. What I would really like is to make a buffer the exact size of the image, that I can blit to the screen, instead of having to perform translations on every pixel as I'm adding then to the image to account for its position on the screen. I should've said I want to hold an image the size of the screen or smaller 03.32.17 # JdGordon: :P 03.32.54 # JdGordon: No need to rush, the statusbar still flickers (in case you were talking about the pitchscreen_vp patch) 03.34.07 Quit dabujo_ (Read error: 54 (Connection reset by peer)) 03.34.12 # num1: something like extern fb_data lcd_framebuffer[LCD_FBHEIGHT][LCD_FBWIDTH]; 03.34.38 # num1: depends on the feauter it may be better to allocate the buffer from the audiobuffer at boot time 03.34.50 # audiobuf_alloc() (or somethying like that) 03.35.07 # not the same ass buffer_alloc() 03.35.23 Quit ChristopherW (Read error: 110 (Connection timed out)) 03.35.37 # JdGordon: buffer_alloc() allocates data in the plugin buffer? 03.35.37 # sorry, got them backwards... use buffer_alloc() 03.36.27 # buffer_alloc() steals some of the audio buffer 03.37.01 # JdGordon: alright, thanks. I was hoping there was a way to dynamically allocate memory without stopping playback 03.38.22 # there is 03.38.34 # give me a min... im putting up a wiki page with the options 03.38.45 Quit nplus (Remote closed the connection) 03.39.16 # if you want to keep the buffer for the whole rockbox "session" then you should either statially allocate it, or use buffer_alloc 03.39.20 Quit DerDome (Read error: 110 (Connection timed out)) 03.44.47 Join ackers [0] (n=Becks@80-235-149-109.cable.ubr22.newt.blueyonder.co.uk) 03.47.35 Quit fdinel (Read error: 110 (Connection timed out)) 03.48.08 Part ackers 03.52.17 Quit Thundercloud (Remote closed the connection) 03.53.50 Quit tehpaper (Read error: 104 (Connection reset by peer)) 03.56.46 Join tehpaper [0] (i=UPP@653237hfc238.tampabay.res.rr.com) 03.59.54 Quit kugel ("ChatZilla 0.9.83 [Firefox 3.0.1/2008072820]") 04.07.25 # arg, the wiki has slowed to a crawl :( 04.07.49 # JdGordon: can you post a link to the page when you're done? 04.09.37 # num1: http://www.rockbox.org/twiki/bin/view/Main/AllocatingRAM 04.09.57 # JdGordon: thank you kindly 04.19.56 # I'm having trouble attaching a files to flyspray reports and comments. And ideas? 04.21.00 # nevermind I found out why. How do I add a patch that's bigger than 2MB? 04.21.11 # s/add/submit 04.21.28 # zip it? 04.21.33 # what file is it? 04.21.35 Nick JdGordon is now known as JdGordon_ (n=jonno@rockbox/developer/JdGordon) 04.21.39 Nick JdGordon_ is now known as JdGordon (n=jonno@rockbox/developer/JdGordon) 04.22.46 # #9235 04.23.36 # can someone look at FS#9234 please? 04.24.40 # yeah, i saw it... its a pretty low priority fix... 04.25.05 # JdGordon: well yes... but it's also a very easy one-line fix 04.29.10 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-a46b1181e2413ffd) 04.33.14 Quit m0f0x ("sleep") 04.34.31 Join miepchen^schlaf_ [0] (n=miepchen@p54BF481E.dip.t-dialin.net) 04.36.04 Join snoh [0] (n=dave@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com) 04.36.06 Quit miepchen^schlaf (Read error: 60 (Operation timed out)) 04.39.22 # JdGordon: I'm assuming that if you're a plugin you can't allocate memory from your plugin buffer at runtime since you didn't add it to the list? 04.41.20 # if your a plugin just use the static buffer... its easiest 04.41.25 # unless its larger than 512K 04.42.05 # what's the static buffer? 04.42.46 # just add a global variable to the plugin 04.42.55 # char mybuffer[BUF_ZISE]; 04.44.05 # JdGordon: alright, thanks 04.45.12 # is vfat.h a standard unix library? 04.45.32 # dunno 04.46.05 # im guessing no though 04.46.52 # sorry vfs 05.02.20 Join solexx [0] (n=jrschulz@e176124200.adsl.alicedsl.de) 05.10.31 *** Saving seen data "./dancer.seen" 05.14.31 # JdGordon: are include files specific to a particular gcc build, or do all the gcc versions (arm, coldfire, etc) use the same include files? 05.14.51 Join cynic [0] (n=cynic@pool-68-160-43-36.bos.east.verizon.net) 05.15.01 Quit cynic (Client Quit) 05.15.30 # saratoga: umm.. i assume they all use the same ones, except where they have arch specific stuff.. but then they would still have the same names, or #included in a common .h? 05.15.57 Quit snoh ("GNU/Linux the free alternative to Windows") 05.15.58 Join cynic [0] (n=cynic@pool-68-160-43-36.bos.east.verizon.net) 05.16.03 Quit cynic (Client Quit) 05.17.15 # JdGordon: so all gcc crosscompilers are still looking at /usr/include? 05.18.32 Join cynic [0] (n=cynic@pool-68-160-43-36.bos.east.verizon.net) 05.18.49 Part cynic ("Leaving") 05.22.28 Quit solexx_ (Read error: 110 (Connection timed out)) 05.32.16 # i've just found the strangest gcc issue 05.32.49 # removing #ifndef _ANSIDECL_H_ from _ansi.h fixes a compilation problem, even though putting a #warning in the file shows that it was not defined 05.33.17 Join HETROSEXO [0] (n=40b41555@gateway/web/cgi-irc/labb.contactor.se/x-bad592a48cb1d419) 05.33.35 Quit HETROSEXO (Client Quit) 05.33.41 Join HETROSEXO [0] (n=40b41555@gateway/web/cgi-irc/labb.contactor.se/x-dffe2577de5d8179) 05.33.50 # WASSUPP 05.34.56 Quit HETROSEXO (Client Quit) 05.36.34 Quit sacarasc (Read error: 110 (Connection timed out)) 05.37.25 Join toffe82 [0] (n=pocketir@189.155.27.171) 05.44.02 Join Llorean [0] (n=DarkkOne@12.187.237.194) 05.52.14 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 06.03.18 Quit sacarasc (Connection reset by peer) 06.04.00 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 06.04.43 Join maha [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 06.05.01 # whats 'caption backlight' ? 06.05.17 # the backlight turns on for a few seconds at the track change 06.05.37 # so whats 'party mode' 06.06.01 # its not very intuitive 06.06.07 # http://www.rockbox.org/manual.shtml 06.08.23 # i hasve always been havin this battery drain issue 06.08.40 # it can do perhaps 10 h 06.08.43 # e200 sansa 06.08.48 # is this normal ? 06.12.19 # i get around 15 to 17 on my sansa.. how old is your sansa 06.13.40 # my e280 gets about 19 depending on how use it 06.13.54 # on how *I* use it 06.16.08 # that sucks 06.16.16 # its like 6 months 06.16.22 # rarely use it actually 06.16.38 # did you batterybench it ? 06.20.27 # i think i did at some point 06.23.54 Quit sitwon (Remote closed the connection) 06.24.26 Quit saratoga ("CGI:IRC (EOF)") 06.25.06 Join sitwon [0] (n=adam@pool-71-178-171-238.washdc.fios.verizon.net) 06.28.17 Join madx [0] (i=Lx@212.57.158.24) 06.28.28 Quit _Lx_ (Read error: 113 (No route to host)) 06.30.09 Quit toffe82 (Remote closed the connection) 07.04.20 Join vitja [0] (n=vitja@79.120.98.174) 07.04.34 # linuxstb, I did it 07.04.51 Quit vort3x (Remote closed the connection) 07.10.33 *** Saving seen data "./dancer.seen" 07.17.36 Join vort3x [0] (n=vortex@unaffiliated/dfa001) 07.18.51 Join ChristopherW [0] (n=christop@ip68-3-205-51.ph.ph.cox.net) 07.28.26 Quit XavierGr (Read error: 110 (Connection timed out)) 07.35.38 # for 16bit color lcd targets, how do I figure out what the format is? Like how are red green and blue packed into those 6 bits? 07.35.46 # s/6/16 07.37.46 # RGBPACK_LCD seems like the answer but it confuses me because it doesn't seem to care what the LCD_DEPTH is 07.42.18 Quit CaptainKewl (Read error: 110 (Connection timed out)) 07.44.15 # nevermind, I found it. 07.44.26 # * num1 tells himself to RTFM before asking questions 07.56.34 Nick madx is now known as ____lx (i=Lx@212.57.158.24) 08.14.35 Nick maha is now known as Sarajebo (i=juma@gateway/gpg-tor/key-0xB9002659) 08.19.34 Nick Sarajebo is now known as maha (i=juma@gateway/gpg-tor/key-0xB9002659) 08.23.06 Quit ____lx ("Leaving") 08.24.09 Nick num1 is now known as num1_sleeping (n=brian@c-68-63-193-45.hsd1.az.comcast.net) 08.28.29 Join Rob2222 [0] (n=Miranda@p4FDCDE1C.dip.t-dialin.net) 08.44.39 Quit Acksaw ("Erection reset by queer") 08.46.21 Quit Rob2223 (Read error: 110 (Connection timed out)) 08.55.20 Join Acksaw [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) 08.55.31 # vitja: What did you do? 08.56.02 # linuxstb, I've written button and simple lcd driver 08.56.15 # and rockbox bootloader booted 08.56.50 # now I'm going to play with usb 2.0 host 08.56.59 # s/host/device/ 08.57.08 # never did it before, only host 08.57.36 # Have you tested the NAND driver? 08.58.46 # when I call ata_init() bootloader panics with nand cache error 08.59.35 # does it work ok on other tcc77x ports? 09.00.11 # It apparently works a little... At least, I haven't heard of that error before. 09.01.41 # ok so.. if run ata_init() in bootloader instead of show_debug_screen() that should work? 09.02.47 # Yes. Are you also calling the other _init() functions? 09.03.30 # here are I use default loader that could be found in bootloader/telechips.c 09.03.54 # I see cowond2 also calls kernel_init() and enable_irq() 09.04.00 # does that makes sens? 09.05.20 # No, the D2 doesn't call those two functions. 09.05.40 # yeah 09.05.46 # I was wrong here 09.06.51 # so I simply enable D2 nand initialization code, and that sais Max NAND write cac has reached 09.08.38 # Is that write? 09.08.46 # I mean right 09.10.36 *** Saving seen data "./dancer.seen" 09.11.03 # Well, something is obviously wrong. shotofadds is the person to talk to about the NAND driver, but I think he may be on vacation for the next couple of weeks... 09.16.03 Quit lordpil (Read error: 104 (Connection reset by peer)) 09.19.57 # my player is 8Gb, if that matter 09.21.26 Join goffa_ [0] (n=goffa@216.220.23.105) 09.22.45 # Have you tried increasing the number of write caches (MAX_WRITE_CACHES in ata-nand-telechips.c) ? 09.23.04 # I've no idea if that's the right thing to do, or if that error is because there is another issue... 09.24.46 # no 09.25.01 # but it seems driver detects my flash correctly 09.26.44 # linuxstb: I set it to 64, and ata_init and mount finish successfully) 09.27.34 # I think that's good... You don't want to set it higher than needed though. 09.27.57 # that was just for try 09.28.33 Quit Acksaw (Read error: 104 (Connection reset by peer)) 09.28.50 # min value it works is 32 09.29.15 # I did't tried non pow of 2 values 09.30.01 Quit k4y_ (Read error: 110 (Connection timed out)) 09.31.21 # hmm... has anyone tried running in the sim in osx? 09.31.35 Join Acksaw [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) 09.31.50 # JdGordon: I have in the past, but not recently... 09.32.30 # it doesnt work anymore.... it loads but becomes unreposive... I have to force quit it 09.32.52 # Has it worked for you previously? 09.32.59 Quit goffa (Read error: 110 (Connection timed out)) 09.33.21 # this is the first time I'm trying in at least 6 months... checking an old bug report 09.33.47 # terminal says we are using an old api and should use Quartz instead of QuickDRaw 09.34.07 # Maybe update the libSDL then? 09.37.37 Quit Acksaw (Read error: 104 (Connection reset by peer)) 09.45.10 Join Acksaw [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) 09.47.10 Nick Horscht86 is now known as Horscht (n=Horscht@p4FD4CBD8.dip.t-dialin.net) 10.00.33 Join stooo [0] (n=sto@y066163.dynamic.ppp.asahi-net.or.jp) 10.00.50 Part stooo 10.20.35 Quit culture (Connection timed out) 10.26.36 Join Schmogel [0] (n=Miranda@p3EE2208D.dip0.t-ipconnect.de) 10.31.04 Quit Schmogel (Read error: 104 (Connection reset by peer)) 10.39.09 Join Mathiasdm [0] (n=Mathias@vpnc137.ugent.be) 10.42.40 Join snoh [0] (n=dave@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com) 10.44.09 Join OlivierBorowski [0] (n=OlivierB@sab57-5-88-179-80-205.fbx.proxad.net) 10.45.34 Join ender` [0] (i=krneki@foo.eternallybored.org) 10.48.06 Join d-k-t [0] (n=dt@125.118.37.84) 10.49.16 Join _Lx_ [0] (i=Lx@212.57.158.24) 10.49.23 Nick _Lx_ is now known as ____Lx_ (i=Lx@212.57.158.24) 11.01.04 Join herrwaldo [0] (n=waldo@ip-81-11-202-47.dsl.scarlet.be) 11.04.40 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk) 11.10.37 *** Saving seen data "./dancer.seen" 11.13.10 Quit d-k-t ("Leaving") 11.25.11 Join culture [0] (n=none@cpc1-bele3-0-0-cust658.belf.cable.ntl.com) 11.25.39 Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl) 11.39.21 Quit Galois (Excess Flood) 11.43.20 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky) 11.43.49 Quit Acksaw (Read error: 104 (Connection reset by peer)) 11.47.18 Join Acksaw [0] (n=omgwtfbb@cpc2-stok5-0-0-cust754.bagu.cable.ntl.com) 11.49.39 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul) 11.50.40 Quit BigBambi (Remote closed the connection) 11.51.20 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi) 11.52.11 Join AndyIL [0] (n=pasha_in@212.14.205.32) 11.59.39 Quit sacarasc (Read error: 104 (Connection reset by peer)) 12.01.54 Join desowin [0] (n=desowin@atheme/member/desowin) 12.03.25 Quit AndyI (Read error: 110 (Connection timed out)) 12.14.29 # Anyone interested in some c200/e200 radio testing? 12.14.52 # e200 testing I mean 12.16.01 # bertrik, what kind of radio testing? 12.17.08 # advcomp2019, some people have trouble with the radio disappearing from the root menu. I'd like to try some things to see if it makes the problem worse or better 12.17.19 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 12.19.39 # i have not ran into that yet myself but i have had another radio issue tho 12.19.55 Quit Horscht ("IRC is just multiplayer notepad") 12.20.56 Join gregzx [0] (n=chatzill@drn194.neoplus.adsl.tpnet.pl) 12.21.31 Quit sacarasc (Read error: 60 (Operation timed out)) 12.21.50 Join goffa [0] (n=goffa@216.220.23.105) 12.24.30 # bertrik, the issue that i have with the radio is they are sometimes off frequency by 0.05 or 0.10MHz 12.25.49 # advcomp2019, I'm not looking into that right now 12.26.49 # yea but i could try using my e200 and see if i can help you out tho 12.33.48 Quit goffa_ (Read error: 110 (Connection timed out)) 12.34.28 # advcomp2019, can you try out this patch and see if it changes anything w.r.t. fm radio detection: http://pastebin.ca/1091677 12.35.44 Join Horscht [0] (n=Horscht@xbmc/user/horscht) 12.36.55 Join desowin_ [0] (n=desowin@atheme/member/desowin) 12.38.23 Quit Horscht (Client Quit) 12.39.34 # ok sorry.. i did not know it was going to be something with building a build.. i have not tried a putting a patch in place.. i should read up a bit more 12.43.30 Join Horscht [0] (n=Horscht@xbmc/user/horscht) 12.44.37 # is there a way to remove the peakmeter totally? cant see it mentioned in the theme 12.46.04 # Just don't include the peakmeter tag. 12.46.59 Quit Seed (Read error: 110 (Connection timed out)) 12.49.24 # linuxstb: yes but its not there in any of the themes I've looked 12.49.29 # advcomp2019, ok no problem. BTW I do have some ideas about improving radio tuning accuracy, but I don't know if it makes a lot of difference 12.49.33 # yet its there annoying me on the playback screen 12.50.16 # I think there's something wrong in my theme setup.. the playback screen doesnt even change as I change the theme 12.51.48 # Ave, that's a common problem, something changed in the WPS syntax, see also http://forums.rockbox.org/index.php?topic=17507.0 12.52.46 # oh ic, well I had a problem in nighty build so I did my own svn update and built from there 12.53.18 # oh noes, ok thanks 13.00.19 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 13.00.24 Quit Horscht ("http://www.geisterfahrer.org") 13.08.54 # turns out %pb definition was and is the problem 13.10.38 *** Saving seen data "./dancer.seen" 13.12.04 Join mcuelenaere [0] (n=mcuelena@78-21-191-98.access.telenet.be) 13.17.09 Quit sacarasc (Read error: 104 (Connection reset by peer)) 13.18.45 Join funman [0] (n=fun@82-171-216-191.ip.telfort.nl) 13.19.22 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 13.25.48 # bertrik: I don't use my e200/c200 often enough to have noticed if the radio periodically disappears, but if you would like I can test patches to see if they work etc 13.34.24 Join Horscht [0] (n=Horscht@xbmc/user/horscht) 13.39.46 Join nplus [0] (n=nplus@141.25.Globcom.Net) 13.40.36 Quit sitwon (Read error: 113 (No route to host)) 13.44.32 Join fdinel [0] (n=Miranda@modemcable204.232-203-24.mc.videotron.ca) 13.47.04 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 13.48.00 Quit sacarasc (Read error: 104 (Connection reset by peer)) 13.52.47 Nick sacarasc_ is now known as sacarasc (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 14.01.39 Join Thundercloud [0] (n=thunderc@84-51-130-71.judith186.adsl.metronet.co.uk) 14.05.38 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 14.06.14 Quit sacarasc (Read error: 104 (Connection reset by peer)) 14.07.22 Join Lear [0] (i=chatzill@rockbox/developer/lear) 14.11.16 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 14.20.25 Join DerDome [0] (n=DerDome@dslb-082-083-244-138.pools.arcor-ip.net) 14.23.34 Quit sacarasc_ (Connection timed out) 14.23.52 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 14.25.21 Quit sacarasc (Connection reset by peer) 14.30.16 Nick sacarasc_ is now known as sacarasc (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 14.31.32 Join petur [50] (n=petur@rockbox/developer/petur) 14.41.51 # BigBambi, thanks for the offer. I never experienced a radio problem myself on my c200 (but I also didn't always check very carefully). I looked at the code yesterday and found some suspicious looking things, but after a closer look they look quite harmless. 14.42.57 # OK, well let me know if you do 14.53.44 Quit sacarasc (Read error: 104 (Connection reset by peer)) 15.09.38 Join ZincAlloy [0] (n=d9eeea98@gateway/web/cgi-irc/labb.contactor.se/x-2cf1d5f0dddbbb6d) 15.10.42 *** Saving seen data "./dancer.seen" 15.10.51 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 15.11.04 Join stooo [0] (n=sto@y066163.dynamic.ppp.asahi-net.or.jp) 15.11.20 Part stooo 15.18.21 # Anyone know what "unported" means/might mean in the following "Creative Commons Attribution - Share Alike 3.0 Unported license" 15.22.52 Join robin0800 [0] (n=robin080@cpc2-brig8-0-0-cust394.brig.cable.ntl.com) 15.24.06 Quit herrwaldo (Remote closed the connection) 15.24.55 Join herrwaldo [0] (n=waldo@ip-81-11-202-47.dsl.scarlet.be) 15.26.12 Quit perrikwp ("http://www.mibbit.com ajax IRC Client") 15.26.30 Quit herrwaldo (Client Quit) 15.26.49 Join funman_ [0] (n=fun@82-171-216-191.ip.telfort.nl) 15.29.13 Join herrwaldo [0] (n=waldo@ip-81-11-220-176.dsl.scarlet.be) 15.32.38 Quit ZincAlloy ("CGI:IRC (Ping timeout)") 15.33.50 Join mf0102 [0] (n=michi@85.127.180.61) 15.38.33 Join LambdaCalculus37 [0] (i=1800dac6@gateway/web/ajax/mibbit.com/x-b575fa96bf19a13d) 15.39.16 Quit herrwaldo (Remote closed the connection) 15.39.45 Join herrwaldo [0] (n=waldo@ip-81-11-220-176.dsl.scarlet.be) 15.40.26 # In arm922t description I see that it has 2 coprocessors: cp14 & cp15 15.40.54 # are they shortened to 'p14' and 'p15' are thos are different coprocessors ? 15.41.04 # or are those* 15.42.21 # well it seem they are just numbered, the prefix means CoProcessor or Processor 15.42.30 # * domonoky thinks "unported" in this context means that its the generic international version, and not a version for a specifc country/language 15.44.42 # domonoky: Could well be :) 15.45.01 Quit sacarasc (Read error: 104 (Connection reset by peer)) 15.45.25 Quit herrwaldo ("Konversation terminated!") 15.45.33 Quit funman (Read error: 110 (Connection timed out)) 15.49.22 Join herrwaldo [0] (n=waldo@ip-81-11-214-184.dsl.scarlet.be) 15.51.31 # is rb.org down? 15.51.53 # The main site is working here. 15.51.53 # works here 15.53.19 Join sacarasc [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 15.54.03 # * JdGordon cant get thourgh :( 16.00.44 Quit funman_ ("Lost terminal") 16.01.01 Join funman [0] (n=fun@82-171-216-191.ip.telfort.nl) 16.02.19 # recovery mode on sansav2 is obtained by briding the WP# pin of the NAND flash 16.02.45 # unfortunately on the Clip I don't see where it could be bridged :/ 16.03.14 # http://www.anythingbutipod.com/archives/images/sansa-clip-disassembled/sandisk-sansa-clip-disassembled-13.jpg < clip's nand flash 16.04.23 Join ZincAlloy [0] (n=d9eeea98@gateway/web/cgi-irc/labb.contactor.se/x-768261b112826459) 16.04.39 Join Galois [0] (i=djao@efnet-math.org) 16.04.40 # oh, I thought most flash chips were pretty similar 16.05.11 # yes but then they can do weird things on the pcb's circuit 16.05.38 # the standard says "when cleared to zero, the WP# signal disable flash array program and erase operations" 16.06.14 # I find strange on the e200 (http://flickr.com/photos/90053035@N00/2495460818/in/set-72157605072639496/) you can get the desired effect by bridging it (that would mean the signal is 1, no ?) 16.06.27 # but you can access all flash chip pins, right, why would you need to know what happens on the pcb? 16.06.53 # "can access", I could if my hands were 10 times smaller ;) 16.07.12 # I do not have any electric equipment 16.09.16 # do we have the data sheet of the flash chip? 16.09.49 Join kugel [0] (n=chatzill@unaffiliated/kugel) 16.10.08 # bertrik: I believe (hope) they use the common wiring 16.10.08 # ONFI 16.10.08 # 48 pins 16.11.57 # JdGordon: Someone uploaded a new version of the pf patch 16.12.46 # oh goody 16.13.02 # if my net didnt suck so much I'd check it out 16.13.11 # He said he fixed your issue 16.13.51 # i see... 16.15.33 Join tvelocity [0] (n=tony@athedsl-144835.home.otenet.gr) 16.15.36 # funman, maybe you can just short it to a pin immediately next to it 16.15.51 # kugel: my dns seems to be fubar so can you pastebin the updated patch for me? 16.16.02 # BigBambi: I think "unported" means it's not ported to any particular jurustiction. 16.16.42 # JdGordon: http://pastebin.ca/1091781 16.16.46 # linuxstb: Cheers - domonoky thought that too, so I'll go with that 16.17.07 # Ah yes, I missed his reply... 16.17.21 # it's next to write enable signal 16.19.52 # yeah, maybe a bit dangerous 16.20.25 # kugel: :) its working... now what were the changes? 16.20.35 # the other one is VSP3, an unspecified vendor specific signal, nothing may be connected to it 16.21.08 # can we figure out exactly which pins are shorted in the e280v2 pic? 16.21.41 # JdGordon: great! 16.21.58 # kugel: actually.. if you have the bmp resize patch in there... why not resize the cover for the track listing screen if there is room? :) 16.22.18 Join CaptainKewl [0] (n=jason@207-237-173-165.c3-0.nyr-ubr4.nyr.ny.cable.rcn.com) 16.22.20 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother) 16.22.28 # hmm.. no it didnt work so well... almost none of my covers is showing 16.23.01 # Llorean: the forums don't use the same favicon as the main website -- can we change that? 16.23.53 # JdGordon: is every of your covers 100x100 16.24.47 # yes 16.25.06 # hm 16.25.42 Quit ompaul (Client Quit) 16.27.23 # kugel: http://pastebin.ca/1091783 16.28.35 # bluebrother: Should be able to, yes. I'll look into it when I get home. 16.29.16 # Llorean: nice :) 16.33.33 Join funman_ [0] (n=fun@82-171-216-191.ip.telfort.nl) 16.35.04 Part Llorean 16.35.10 # woohoo I got a special mode on the clip 16.35.10 # sadly it shows a 0mb hard disk 16.35.10 # I had to short together PIN 17 & 18 of the NAND 16.35.13 Quit funman (Nick collision from services.) 16.35.23 Nick funman_ is now known as funman (n=fun@82-171-216-191.ip.telfort.nl) 16.36.09 # funman: It's progress. :) 16.36.55 # I suppose sansa didn't want to use this feature, and so didn't allocate any room for the recovery mode 16.38.36 # JdGordon: Is the database properly initialized? 16.39.05 # shouold be 16.39.23 # yep 16.40.13 Quit LambdaCalculus37 ("http://www.mibbit.com ajax IRC Client") 16.43.05 # JdGordon: Regarding the resizing for tracklist. Currently, I only implemented the resize algorithm for initializing-time. The resizing which is used for zoom in/zoom out(at selecting an album) or the zoom option in the settings is another algorithm 16.44.11 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 16.44.53 Quit sacarasc (Read error: 104 (Connection reset by peer)) 16.46.52 # it works also by shorting IO pins 16.47.25 # funman, so it's probably anything that makes the crc over the data read by the sansa bl fail? 16.47.34 Join Nibbler [0] (n=Nibbler@port-83-236-56-2.dynamic.qsc.de) 16.48.20 # or maybe anything that makes the flash id fail 16.48.46 # JdGordon: weird, that it doesn't find all album art 16.49.44 # bertrik: from what I read, it's a feature included in the SoC 16.49.59 # it will use its own bootloader if it fails to read from nand 16.50.35 # see http://forums.rockbox.org/index.php?topic=14064.msg127675#msg127675 16.50.45 # JdGordon: What happens if you turn of resizing in the pf settings? 16.51.01 # (and reinitialize) 16.51.13 # maybe the firmware updater of sansa recognize this device and issues commands to it to flash a correct firmware 16.51.33 # kugel: no change 16.51.40 # * JdGordon going to bed 16.51.43 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon) 16.51.47 # I'll try that (I use virtualbox on linux to do usb device pass through till the virtual windows xp) 16.52.53 # * bertrik finds another inconsistency between voiced data and data shown on screen 16.54.58 Quit ComradeHaz (Read error: 104 (Connection reset by peer)) 17.08.51 Quit tvelocity (Read error: 110 (Connection timed out)) 17.09.12 Join tvelocity [0] (n=tony@athedsl-416126.home.otenet.gr) 17.10.44 *** Saving seen data "./dancer.seen" 17.10.53 # Who are the wiki admins? 17.13.43 Nick sacarasc_ is now known as sacarasc (i=sacarasc@xmms2/bitch/sacarasc) 17.15.18 Join sitwon [0] (n=adam@pool-71-178-171-238.washdc.fios.verizon.net) 17.16.27 Quit gregzx ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]") 17.19.08 # Soap: The Swedes I believe 17.19.50 Join mf0102_ [0] (n=michi@85.127.180.61) 17.22.18 Join funman_ [0] (n=fun@82-171-216-191.ip.telfort.nl) 17.22.48 Quit funman (Read error: 54 (Connection reset by peer)) 17.23.46 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey) 17.24.50 Nick funman_ is now known as funman (n=fun@82-171-216-191.ip.telfort.nl) 17.24.52 # what convention should rockbox use w.r.t. mebibytes vs. megabytes? 17.26.01 Join tobbie [0] (i=tobbie@host206-66-dynamic.10-79-r.retail.telecomitalia.it) 17.26.15 # j 17.26.26 Quit mf0102 (Read error: 110 (Connection timed out)) 17.26.36 # y 17.26.37 # y 17.27.21 Part tobbie 17.28.04 Join tobbie [0] (i=tobbie@host206-66-dynamic.10-79-r.retail.telecomitalia.it) 17.32.38 Part tobbie 17.39.21 Quit stripwax (Read error: 104 (Connection reset by peer)) 17.59.10 Quit Nibbler (Read error: 113 (No route to host)) 17.59.18 Quit CaptainKewl (Remote closed the connection) 18.03.49 Quit rvvs89 (Read error: 60 (Operation timed out)) 18.05.43 Join rvvs89 [0] (n=rvvs89@bright-snat.ucc.asn.au) 18.11.06 Join Cyberdyskinesia [0] (n=lachienn@AStDenis-107-1-46-213.w90-43.abo.wanadoo.fr) 18.12.06 Join coatman [0] (n=Jacob@ppp-70-246-143-150.dsl.stlsmo.swbell.net) 18.13.09 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 18.14.05 Quit sacarasc (Read error: 104 (Connection reset by peer)) 18.16.35 Nick sacarasc_ is now known as sacarasc (i=sacarasc@xmms2/bitch/sacarasc) 18.17.31 Join Zarggg [0] (n=z@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com) 18.18.55 Join amiconn [0] (i=4fd6f98c@gateway/web/ajax/mibbit.com/x-93deb64afd53426a) 18.24.14 Quit nplus (Remote closed the connection) 18.25.02 Join nplus [0] (n=nplus@141.25.Globcom.Net) 18.29.22 Quit avis (Remote closed the connection) 18.31.22 # bertrik: Why does rockpaint not simply use lcd_drawline() ? 18.33.43 Quit Cyberdyskinesia ("Leaving") 18.34.05 Quit desowin_ (Read error: 113 (No route to host)) 18.37.23 Nick num1_sleeping is now known as num1 (n=brian@c-68-63-193-45.hsd1.az.comcast.net) 18.40.29 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/mibbit.com/x-02a526ada469a024) 18.41.46 Quit coatman (Read error: 104 (Connection reset by peer)) 18.43.21 Quit funman ("Lost terminal") 18.43.39 Quit nplus (Read error: 110 (Connection timed out)) 18.47.32 Join funman [0] (n=fun@82-171-216-191.ip.telfort.nl) 18.54.46 Join nplus [0] (n=nplus@141.25.Globcom.Net) 18.54.54 Join Seed [0] (i=ben@bzq-84-108-237-178.cablep.bezeqint.net) 18.55.23 # amiconn: Looks like it does use lcd_drawline in places.... But it seems to keep it's own copy of the LCD contents in a "save_buffer", so is always drawing in two places. 18.56.01 # Sounds... unnecessary 18.59.41 Quit robin0800 (Read error: 104 (Connection reset by peer)) 19.04.38 Join desowin_ [0] (n=desowin@atheme/member/desowin) 19.05.03 Join Horschti [0] (n=Horscht@p4FD4F832.dip.t-dialin.net) 19.05.47 Quit Horscht (Nick collision from services.) 19.07.28 Quit Seed (Nick collision from services.) 19.07.56 Join Seed [0] (n=ben@bzq-84-108-237-178.cablep.bezeqint.net) 19.10.37 # amiconn, I was not aware of that function, but in hindsight it sounds logical that it would be available 19.10.47 *** Saving seen data "./dancer.seen" 19.13.37 # I wonder why rockpaint needs that backbuffer 19.14.00 Nick num1 is now known as num1_away (n=brian@c-68-63-193-45.hsd1.az.comcast.net) 19.14.06 # I know it saves .bmp as well as loading it, but then maybe the screendump function could be used for this 19.15.34 # It seems to call a "restore_screen" function many times - after drawing UI things on top of the image. 19.15.51 # hmm 19.16.22 # Which simply uses lcd_bitmap() to draw a full-screen bitmap (the save_buffer[]) 19.16.35 Quit funman ("leaving") 19.16.47 # It may be necessary to keep the backbuffer then. I'd probably go for simple memcpy()ing in this case... 19.17.51 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net) 19.22.06 Join goffa_ [0] (n=goffa@216.220.23.105) 19.24.10 # amiconn: Do you have a view on the bitmap resize patch in flyspray? (FS#5697). It basically adds the following function to the core (in apps/recorder/resize.[ch]): void resize_bitmap(struct bitmap *src, struct bitmap *dst) 19.25.43 # grr, clicking a FS# in mibbit has an unwanted effect... 19.25.57 # It opens a new channel (because of the hash sign) 19.27.58 Quit desowin_ ("KVIrc 4.0.0 Insomnia http://www.kvirc.net/") 19.28.49 # Well, the patch seems to do smooth resizing now. Can't say much as I don't have the possibility to test it right now, and there are lots of comments... 19.30.58 # I do agree that we need resizing in the core, and in fact it should imo have all of the following features: (1) Be smooth (i.e. no simple pixel dropping/ duplicating) (2) Be fast and compact. (3) Be available for all pixel formats which support album art (i.e. all depths >= 2) 19.33.48 # (3) shouldn't be difficult if it's done on the intermediate line buffer (that buffer is alwas 32 bit after unifying input data, before converting to the output format) 19.33.54 Quit goffa (Read error: 110 (Connection timed out)) 19.38.13 # It may be necessary to keep 2 intermediate buffers though (for averaging), and I'm not sure whether it's a good idea to keep them on the stack then (right now it is, but that's already ~1.3KB for the target with the widest screen (ipod video), and will become worse with some new trgets (m:robe 500, Onda)) 19.39.28 # Are you talking about resizing-on-load - in the read_bmp() function? 19.39.34 # yes 19.40.04 # Because that's not what the patch does... It implements a resize_bitmap(src,dst) function. 19.40.20 # Hmm. Imho that's the wrong way 19.41.31 Quit Mathiasdm ("Invisible Internet Project: http://www.i2p2.de") 19.41.53 # I think the main purpose is scaling down, and then you waste lots of ram if you resize after loading 19.41.53 # Yes, that's my view as well. 19.42.32 # A good "live" resizing algorithm is what we need 19.43.00 # btw, I want one for the jpeg viewer as well, to do away with wasting screen space in the default view 19.43.22 # I also want (optional) auto rotation for best fit in the jpeg viewer 19.43.23 # The AA-related parts of the patch don't look very reliable - IIUC, it just uses the remaining part of the audio buffer as temporary storage, ignoring wraparound issues. 19.43.59 # hmm 19.44.15 # I will have a closer look tomorrow, when I'm back home 19.44.17 # Nico_P was going to look at the patch and comment, so I'm leaving that to him... 19.44.23 # (the buffering issues) 19.45.08 # If all the mentioned parts are implemented, it may be fairly easy to add jpeg album art support. If we want that, that is 19.45.23 # I read about a smart algorithm (from DDJ IIRC) that first scaled the image down to about the proper size (bigger than target, but less than 2 times) by scaling down in steps of 2 (easy, just average 4 pixels), then did a bilinear step. Because the image is at this point less than twice as big as it needs to be finally, the bilinear algorithm only needs to consider 2 input pixels for each output pixel. 19.46.03 # amiconn: I think the general view is that we do want jpeg album-art support. 19.46.13 Join LambdaCalculus37 [0] (n=LambdaCa@c-24-0-218-198.hsd1.nj.comcast.net) 19.46.59 # bertrik: The idea itself sounds good, but there is a fundamental problem: the scaling algorithm we need needs to work with only 2 (maybe 3, but as few as possible) original lines available at most 19.47.33 # i.e. it will never have access to the complete original image at once 19.47.36 # How does jpeg decoder decode the image? i.e. in what order do the pixels come out? 19.48.14 # It decodes in natural order, i.e. macroblock after macroblock 19.49.02 # But the jpeg viewer *does* buffer whole images (starting with the 1/8 x 1/8 scaled one, or the biggest one that fits the display, whatever is bigger) 19.50.10 # The advantage of using jpeg for album art would be that we don't need extreme scaling factors if we demand a sane limit for the input image, due to the possible "natural" jpeg downscaling 19.51.03 # Scaling by 2^x should be easy to do on-the-fly 19.51.30 # bertrik: It's simple in horizontal direction, but not so simple in vertical direction 19.53.12 # Are we really that restricted w.r.t. buffer space? 19.55.13 # Some targets are... 19.55.25 # On most targets that support album art we aren't, but wasting resources can't be good 19.55.44 # heh now theres an idea 19.55.49 # a snes emu for rockbox 19.56.09 # * LambdaCalculus37 thinks tehpaper is completely bonkers 19.56.34 Quit ZincAlloy ("CGI:IRC (Ping timeout)") 20.02.02 # LambdaCalculus37, tehpaper? 20.02.27 # play snes games on a mp3 player 20.02.55 Part amiconn 20.03.38 # tehpaper: So you plan on emulating a 65C816 CPU, SPC700 chip, and the graphics chip, which is processor intensive already, on limited embedded hardware? 20.03.40 # * bluebrother has an idea: remove doom! 20.03.42 # Good luck with that! 20.04.27 # linuxstb: Regarding the bmp resize, am I wrong when I assume that the smoother the algorithm should resize the more it needs about to know about the image (i.e. the more lines it needs at live resizing before doing anything)? 20.04.33 # all im coming up with is an idea, dont be a dick. how about give reason A B and C why it wont work well 20.04.43 # * kugel hopes he got it 20.05.57 # if you can get a full emu on a nokia phone i cant see how much harder it would be to get it on a ipod 20.06.19 # tehpaper: then start coding :P 20.07.12 # kugel: I guess it would depend on the algorithm. Graphics processing isn't my speciality, so don't ask me... 20.07.49 # tehpaper: nokia phones have about a 400MHz CPU, the ipods Rockbox run on have a 80MHz CPU... (although it's dual-core) 20.08.37 # linuxstb: The algorithm satisfies all of amiconn's demands, the only issue seems to be the live-resizing 20.08.47 # linuxstb: I don't think all nokia phones have a 400Mhz CPU, but at least they're indeed faster than an iPod 20.09.05 # mcuelenaere: Don't spoil my argument ;) 20.09.13 # :) 20.09.17 # The Gigabeat F and S would possibly be able to handle it, but still... 20.09.29 # Not many of the other targets can. 20.09.31 # kugel: No, he said "that's the wrong way". 20.10.26 # as far as I understood, we can't load the image and then resize it 20.10.27 # linuxstb: That's what I said. 20.11.35 # bertrik: as far as I understood, we can, but it's not wanted 20.11.52 # bertrik: the patch basically show that we can 20.12.56 # * bertrik looks up the patch again 20.16.47 Join sacarasc_ [0] (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 20.17.11 Quit sacarasc (Read error: 104 (Connection reset by peer)) 20.17.40 # as far as I understand (it could use some more comments), that patch also needs a big buffer containing the image in its original dimensions 20.18.46 # Yes, the resize function takes a pointer to the input bitmap, and a pointer to the output bitmap. 20.19.12 # so it's out of the question because amiconn indicated we can buffer 2 or 3 lines at most 20.20.16 # s/can/are allowed to/ 20.34.50 Join Mathiasdm [0] (n=Mathias@vpnh248.ugent.be) 20.41.15 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/labb.contactor.se/x-9cf2fd3e8cc51ce5) 20.41.26 # for a bilinear interpolator, theres no sense buffering the entire image anyway 20.41.31 # you only ever need 2 lines 20.44.24 # And another advantage of doing it on load is that it can be done before the pixels are packed into the device's LCD format - so it will be much simpler. 20.47.51 # * bertrik wonders if you can do interpolation with some bresenham-like algorithm 20.48.24 # hmm, speed is not nearly as important as ram and code size though 20.52.52 # bilinear should be about as fast and compact as you're going to get for a 2x2 point interpolation 20.54.21 Nick sacarasc_ is now known as sacarasc (i=sacarasc@cpc1-kemp3-0-0-cust978.lutn.cable.ntl.com) 20.58.12 # but I assume we also need some kind of anti-aliasing 20.59.25 # LambdaCalculus37: The Spanish translation contains some omissions, as can be seen here: http://rasher.dk/rockbox/translate/problems.php?lang=espanol (for example, LANG_UNKNOWN is translated by "Unknown") 20.59.53 # bertrik: anti-aliasing is another word for interpolation 21.00.04 # in this case the AA is a linear interpolant 21.01.41 # rasher: No problem, I'll start looking through it and making some notes on how best to translate. 21.02.03 # One thing we should make sure of is that the Spanish translation is using Castillian Spanish. 21.06.31 # rasher: I'll do some work on it tonight. 21.06.35 # * LambdaCalculus37 has to go for now 21.06.37 Quit LambdaCalculus37 ("Do quit now, there's a demon around the corner!") 21.07.04 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk) 21.10.14 # saratoga, oh no, it's definitely not the same 21.10.51 *** Saving seen data "./dancer.seen" 21.13.58 Quit vitja ("Ex-Chat") 21.16.02 Quit Mathiasdm ("Yuuw!") 21.16.09 # Seems the bmp resize patch used to do it in read_bmp(), but the idak came along and posted his implementation, and no-one asked why.... (unless I missed it). 21.16.33 # * linuxstb thinks 202 comments is a bit excessive for one task... 21.19.37 Join Maha` [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 21.19.37 Quit maha (Nick collision from services.) 21.20.06 Quit OlivierBorowski (Remote closed the connection) 21.20.25 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul) 21.20.28 Join OlivierBorowski [0] (n=OlivierB@sab57-5-88-179-80-205.fbx.proxad.net) 21.21.37 Join Llorean [0] (n=DarkkOne@32.134.163.235) 21.25.01 Quit thegeek (Read error: 104 (Connection reset by peer)) 21.25.34 Quit Bensawsome (Read error: 104 (Connection reset by peer)) 21.26.54 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome) 21.31.14 Join kintesh [0] (n=5cea23d5@gateway/web/cgi-irc/labb.contactor.se/x-2638e383f74bc603) 21.31.22 # hello 21.32.24 # I know a native castilian speaker - do you want me to use him? 21.32.24 Quit |Marco| (Read error: 110 (Connection timed out)) 21.32.36 # dosre it works on my philips mp3 video player 21.32.41 Quit kintesh (Client Quit) 21.33.22 # no 21.34.14 # bluebrother, I'm trying to create dutch voice files in rbutilqt, so I added "-v dutch-test" as TTS options, but I get a segmentation fault. Is this supposed to work yet? 21.36.43 # bertrik: as espeak option? 21.36.58 # that should work afaik. At least it shouldn't segfault ... 21.37.32 # bluebrother, yes, it segfaults in wavtrim when creating voice files 21.37.33 Join MethoS [0] (n=clemens@91.97.240.183) 21.38.01 Join thegeek [0] (n=nnscript@s080a.studby.ntnu.no) 21.38.32 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr) 21.39.48 Quit Maha` (Remote closed the connection) 21.40.08 # strange. I just tried adding the same as espeak tts option. It doesn't work properly but at least it doesn't crash. 21.40.29 # which OS are you using? I'm on linux. 21.42.13 # I'm on ubuntu, calling espeak with option -v dutch-test works fine on the command line 21.42.18 Join maha [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 21.43.30 # oh, it runs better with quotes around it 21.44.13 # preglow: the Norwegian translation seems to be using "" rather than none, almost all over the place. 21.44.44 Join ElChorristo [0] (n=57b6f213@gateway/web/cgi-irc/labb.contactor.se/x-c843e6d4ca06191b) 21.44.49 # hi 21.44.54 # bertrik: quotes around the complete string? 21.45.19 # is here one of the rockbox creators? 21.45.34 # not really ... a bunch of 'em 21.45.36 # quotes around -v dutch-test so it looks like "-v dutch-test" in the input box 21.45.55 # hmm, then that text should get automatically quoted. 21.46.48 # maybe they are intrestet: Theres is a tool to modify the iPod Touch firmware..maybe the can write a rockbox for it now. the tool is called "pwnage (Mac) and winpwn (windows) 21.46.57 # *interested 21.47.23 Quit ElChorristo (Client Quit) 21.48.23 # said and leaves. Wow. 21.49.09 # * mcuelenaere thinks the buildserver is stuck again 21.49.11 # maybe he was pwned? 21.49.50 # Apple got him 21.51.50 Join Mathiasdm [0] (n=Mathias@vpnf078.ugent.be) 21.52.03 Join avis [0] (n=ident@pdpc/supporter/student/avis) 21.52.30 Join zePh7r [0] (n=nome@89.181.54.192) 21.52.52 # hi. I'm trying to submit a new lang file but tracker doesn't seem to be available... 21.53.57 # zePh7r: the index page isn't available, the rest is 21.54.05 # zePh7r: Some pages don't work (known issue), but you should be able to access the patch page directly - http://www.rockbox.org/tracker/index.php?type=4 - and then "Add new task" 21.56.29 # * domonoky does not understand why you would need to quote the options string for espeak, its already quoted in the template.. (see tts.cpp) 22.01.04 Join |Marco| [0] (i=marco@cakebox.net) 22.02.13 Quit maha (Nick collision from services.) 22.02.14 Join Maha` [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 22.03.00 # domonoky: I guess you must *not* quote the options string -- looks like espeak treats the string as a single input if quoted 22.03.03 Quit desowin (Excess Flood) 22.03.23 Join desowin [0] (n=desowin@atheme/member/desowin) 22.03.34 # and I wouldn't be surprised if the other tts' would behave the same. Basically you're grouping all parts of the options into a single string ... 22.04.20 # Hmm, I inputted -v dutch-test, but on the console it shows Failed to read voice ' dutch-test' (with a space as the first character) 22.04.30 # yeps: just tried it. If I run "espeak -v de" from the command line it works as expected. If I run "espeak "-v de"" it still works but the wrong language is chosen. 22.04.44 # hm, thats right.. should remove the quotes for the options in the template :-) 22.05.33 # you only need to make sure the option string is getting passed as whole, but with QStrings there shouldn't be any issue about this ... 22.06.15 # It seems to work without the space, like -vdutch-test or -vnl 22.07.06 # works with space for me too (at least from the command line) 22.07.41 # bluebrother, yes it works here too on the command line, just not in rbutilqt 22.07.45 # can someone test with the quotes removed from the template ? ( i dont have espeak installed at moment).. 22.08.42 # heh: that won't work. QProcess::execute() wants a QStringList for the arguments, so you need to split it up. 22.08.49 # will try, gimme a few minutes. 22.09.34 # no i give the complete string with options and all to execute, no stringList.. 22.10.18 # so just removing the quotes from the template should work.. 22.10.52 # but that's strange: from the docs it's int QProcess::execute ( const QString & program, const QStringList & arguments ) 22.12.01 Nick Maha` is now known as maha (i=juma@gateway/gpg-tor/key-0xB9002659) 22.12.02 # bluebrother: there is a overloaded version which takes the complete string with arguments.. 22.12.21 # also, ignoring the return code of execute() doesn't seem like a good idea to me 22.12.30 # domonoky: can't find this overloaded version in the docs :/ 22.12.49 # http://doc.trolltech.com/4.4/qprocess.html#execute-2 22.12.51 # Maybe it only matters on some OSes.... But on Unix, the arguments should be a list (IIUC) 22.12.54 Nick MethoS is now known as MethoS- (n=clemens@91.97.240.183) 22.12.58 Nick MethoS- is now known as MethoS (n=clemens@91.97.240.183) 22.13.46 # that doesn't tell anything about the arguments -- it only has QString& program 22.14.05 # "program is a single string of text containing both the program name and its arguments. The arguments are separated by one or more spaces." 22.14.07 # ah, now I see 22.14.21 # * bluebrother didn't read far enough 22.14.55 # and checking the return code is difficult, as the TTSExes is a generic Executable TTS class, not all give the same return codes.. 22.15.28 # can't we assume all executables giving the return code 0 if no error occured? 22.15.40 # someday we should use the espeak libarary instead of the exe.. or even a built-in version of espeak.. 22.16.16 # You would need to go GPLv3 for rbutil if you did that though... 22.16.19 Quit ____Lx_ (Read error: 113 (No route to host)) 22.17.07 # yep, removing the quotes helps. 22.17.22 # who wants to commit? 22.17.23 # arg, always this licence problems... 22.17.28 Part Llorean 22.17.32 # domonoky: I don't think it's a problem though. 22.17.59 # sansapatcher/ipodpatcher are already "v2 or later", what is Qt? 22.18.00 Quit zePh7r (Connection timed out) 22.18.07 # bluebrother: go for it.. 22.19.33 # I'm trying to add translations to the turkish language file, but I realized a lot of the translations that were already there had had their characters romanized,( c instead of ç and so forth) is it not possible to use unicode chars in the translations? 22.20.32 # num1_away: Yes, lang files can (and should) be utf-8 22.20.36 # it seems Qt is v2 or later, so should be okey... 22.20.36 # ok, committed. 22.20.38 Nick num1_away is now known as num1 (n=brian@c-68-63-193-45.hsd1.az.comcast.net) 22.20.51 # Maybe those strings date from before Rockbox had unicode support. 22.21.12 # Qt4.4 introduces a GPL3 license option IIRC. 22.21.48 # but I'm not sure if I would want rbutil having a different license than rockbox itself 22.21.50 # How much impact does it have for other languages when I change an english string? 22.22.11 # * domonoky checked a Qt source file, states v2 or v3, any later, approved (from trolltech and KDE) version.. 22.22.36 # how does the tag work in language files? 22.22.39 # bertrik: in rbutil translations? 22.22.47 # s/in/for/ 22.22.57 Quit maha (Nick collision from services.) 22.22.59 # bluebrother, no, for the rockbox firmware 22.23.00 Join Maha` [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 22.23.01 # I mean, where do you put the voice sound files and how do you give it a name for the language file to use? 22.23.04 # bluebrother: I can't see it causing any issues. Rockbox itself is v2 or later, so will always be compatible with rbutil. And individual parts of rbutil could still be "v2 or later", just the whole thing is v3.... 22.23.06 # * linuxstb sighs... 22.23.48 # bertrik: Changing the meaning of a string isn't good. The other languages would show the old meaning until updated. 22.24.27 # num1: the voice tag in language files is uses by our voicegeneration (buildsystem or rbutil) to generate the voiceclips via a tts engine.. . 22.24.59 # Lear, it's not really the meaning that I want to change, but some C format converters, see http://www.rockbox.org/tracker/task/9237 22.25.04 # domonoky: what happens if the tts engine doesn't support a given language? 22.25.30 # linuxstb: true, but I'm not sure if I would be happy with such a split license situation. 22.25.46 # num1: then you should use another tts engine, or it will sound ugly.. :-) 22.26.44 # theoretically it also possible to generate a voice out of human speech, but thats too much work.. 22.26.55 # bertrik: IMO we should use either IEC prefixes or SI prefixes, but always use the correct divider / abbreviation ... 22.27.14 # domonoky: it's not as much work as you'd think. I did it at one point, took about 5-6 hours total to record the samples, but I haven't been updating. 22.27.15 # domonoky: well, if someone wants to do the job ... ;-) 22.27.33 # domonoky: isn't that what tts engines do anyway :D 22.27.57 # num1: Sounds good that you're willing to do the Turkish translation - maybe it's easier to just start over, by editing a copy of english.lang and translating everything in and , since the current turkish translation is VERY outdated 22.28.02 # bluebrother, OK, the patch just tries to be consistent with the current rockbox convention (which may be wrong) 22.28.07 # (and as you say, doesn't properly use utf-8) 22.28.45 # it has been discussed before, but I'm strongly in favor of being correct. 1024 bytes is *not* 1 kilobyte. 22.29.39 # bertrik: Well, not the absolute meaning, but the change is "big", so to speak. In this case, unfixed languages would show something like "Buf: %d.%03dMB 28.1 MB". Not fatal, but not pretty either. :) 22.29.50 # bluebrother: "A kilobyte (derived from the SI prefix kilo-, meaning 1,000) is a unit of information or computer storage equal to either 1,024 bytes (210) or 1,000 bytes (103), depending on context." 22.29.58 # (from wikipedia) 22.30.05 # bluebrother: you've got my support. We should change instances to be 1000-based, or use the binary prefixes, as appropriate (I assume bitrates use powers-of-ten based prefixes). 22.30.17 Nick Maha` is now known as maha (i=juma@gateway/gpg-tor/key-0xB9002659) 22.30.18 # Bagder: IETF, ISO and that other one disagrees. 22.30.23 # Bagder: well, while that is common useage it's technically not correct. 22.30.26 # rasher: allright, I'll do that. Should I just translate and copy strings from dest to voice? Why are there two fields there? 22.30.43 # bluebrother, rasher: you can argue that, people will still disagree 22.30.50 # num1: Because in some cases, the two may not be identical (or either one of them are empty) 22.30.51 # Lear, so *if* I change it, I should do this to all languages, right? 22.30.53 # kilo is 1000 and 1024, face it 22.31.04 # Bagder: Only until we sort out the damage done 22.31.09 # kilo is 1000 and some people think it's 1024, including windows. 22.31.24 # as per *Standard* it's always 1000. 22.31.36 # bluebrother: "people" thought it was 1024 even before windows existed 22.31.50 # rasher: In what cases would they be different? It seems like you want the program to read what is printed on the screen 22.31.51 # in computer land kilo was always 1025 22.31.53 # 1024 22.32.00 # Bagder: Not true either 22.32.02 # until people started to deal with the confusion 22.32.13 # "computer land" is no standard ... 22.32.21 # I learn kilo == 1024 in the 80s 22.32.26 # learned 22.32.26 # it's just some kind of usage. 22.32.29 # Kilo has been 1000, at first, then it became either 1024 or 1000. 22.32.33 # bertrik: That would be nice, and it should be possible to figure out what the "MB" part is, even if you don't speak the language in question. 22.32.36 # bluebrother: Computers weren't invented in the 80s though 22.32.45 # Bagder: It depends on whether you're a hard drive maker/marketer or a programmer :D 22.32.45 Quit maha (Nick collision from services.) 22.32.46 # as usual, xkcd has something to say about it too: http://xkcd.com/394/ :P 22.32.47 Join Maha` [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 22.32.51 Join pixelma [0] (i=pixelma@212.204.47.129) 22.33.01 # num1: Perhaps you want to use an abbreviation on the screen, but spell out the full words 22.33.07 # num1: I'm not convinced they've been consistent either 22.33.09 # Bagder: Except when dealing with mass storage manufacturers... 22.33.13 # num1: Or a scientist... 22.33.15 # Just display it in bytes... 22.33.20 # rasher: lol 22.33.25 # Or CORRECT, for that matter. 22.33.30 # bertrik: okay, thanks. That makes sense 22.33.32 # Which the hard drive makers happen to be. 22.33.35 # well, I really don't care if use kiB or kB, but if we use 1024 as divider it should be kiB. 22.34.07 # KiB? 22.34.10 # * rasher already changed dansk.lang to do this in places where the translation can handle it 22.34.15 # bertrik: 1KiB = 1024 bytes 22.34.18 # * num1 deals with the Intel Kilobyte every day 22.34.35 # yes, I know, but bluebrother called it kiB 22.34.55 # * bluebrother blames the shift key 22.35.06 Quit Mathiasdm ("Invisible Internet Project: http://www.i2p2.de") 22.35.14 # * domonoky wants this KiB = 1024 * sqrt(-i) from kcd.. :-) 22.35.26 # (and I checked - the prefixes in the langfiles are *only* used in powers-of-two based contexts) 22.35.41 # I think it's sqrt(-1) ;-) 22.36.21 # * domonoky wants complex disk sizes.. :-) 22.36.36 # Kilo has consistently meant 1000 for many hundred years. It has never consistently meant 1024, even within the realms of computing, for the past 40 years. Ising kilo to mean 1024 is sloppy and bad. 22.36.59 # Something that could also be said for my typing. 22.37.12 # rasher: It's consistently meant 1024 when dealing with numbers of bytes, at least in my mind... 22.37.45 # besides, the binary prefixes were invented several years ago to end this confusion. Unless we use them the confusion won't get better but only worse, especially for users. 22.37.49 # linuxstb: That's not what reality has to say on the matter. Harddrive makers for one, transmissions speeds for another. 22.38.05 # linuxstb: just think of network speeds -- in that context it was always 1000. 22.38.09 # * num1 thinks defining capitalization to change meaning is retarded and opening the door to misunderstandings 22.38.09 # Transmission speeds are in bits 22.38.46 # linuxstb: do you really think we should base our prefixes on bits vs bytes, and expect everyone to be on the same page? 22.39.13 # Fact is, when you see KiB, you *know* what is meant, the same can't be said for kB 22.39.31 # * bluebrother really doesn't understand why it's so hard to simply switch to IEC prefixes 22.39.39 # it only makes things clearer. 22.39.42 # At least not until people start using the binary prefixes 22.40.05 # Right, IEEE-1541, ISO/IEC-80000 is what I meant 22.40.05 # which is generally a good thing. 22.40.19 # rasher: I didn't say that. I just said that in my mind, MB, KB etc are binary, and have been for the last 25 years since I started using computers. 22.40.38 # linuxstb: But it's not true though. You constantly have to stop and think. 22.40.59 # buy a 80GB harddisk for your dap of choice ... 22.41.08 # On top of using a different definition of scientific terms than what the rest of the world 22.41.17 # * domonoky has no problem with changing to display KiB instead of kB.. changing to /1000 would be bad.. 22.41.23 # MB and kB were always very clear to me until harddisk makers started to cut corners 22.41.24 # I really don't understand why people don't want to fix it 22.41.30 # No I don't. I know the exception of disk manufacturers. 22.41.58 # maybe you know, but how about users? They are constantly confused. 22.41.59 Join _Lx_ [0] (i=Lx@212.57.158.24) 22.42.34 # "Gain" (LANG_GAIN) means the eq/recording volume? 22.42.59 # I really don't see why admitting that we're not using the terms in their *actual meaning* is such a problem, and moving on to use the proper terminology is an even bigger problem 22.43.01 # "help! My 20GB DAP shows only 18.something in explorer!" 22.43.13 # num1: Correct, amplification 22.43.24 # lol @ bluebrother 22.43.42 # num1: unfortunately this isn't uncommon among users. 22.44.07 # "something is eating up my disc space, explorer doesn't show what I've paid for!" 22.44.15 # just screw those users, we dont need them ! *haha* :-) 22.44.34 # * bluebrother hears some evil laughter :o 22.44.39 Quit Maha` (Remote closed the connection) 22.44.40 # the men-in-black confiscated it to spy on you 22.45.17 # bluebrother: I find it somewhat ironic that harddrive makers have been forced to write a fairy tale to compensate for Windows being wrong. "Some space will be used for system files and might not be available for you" 22.45.53 # iirc gnome also shows binary prefixes ... why can't we do? 22.46.07 # * num1 creates some wierd-ass disk format that stores the partition table and such on a *different* disk, and then sells harddrives marketed as not ripping you off and giving you all the space you paid for 22.46.27 # rasher: true. But eating up 1.something GB of a 80GB or so HDD is really much system ;-) 22.46.42 # bluebrother: As far as I can tell, the answer is "everyone understands this", apparently. 22.47.17 # sure ... windows also reservers (amount of memory + a bit) on my disc for hibernating whole the time 22.47.46 # so this can get quite a bit of space. But it doesn't address the issue itself. 22.47.58 # well linux has swapfiles, so this isn't an only windows thing 22.48.15 # I think they mean the partition structures, like the MFT 22.48.43 Join maha [0] (i=juma@gateway/gpg-tor/key-0xB9002659) 22.48.45 # petur: that was the "and such" 22.49.33 # http://en.wikipedia.org/wiki/NTFS#Metafiles 22.49.39 # (for windows) 22.49.41 # petur: I don't think they mean anything in particular. They just have to put *something* there. Remember, they were sued (and lost, though I don't know where) for using scientific terms correctly (!) 22.51.25 Join jhulst [0] (n=jhulst@unaffiliated/jhulst) 22.51.56 # I've seen newer hard drives having an explanation like "GB means 1 millin bytes" printed on them 22.52.06 # /millin/million/ 22.53.09 # still, I'd really like to see Rockbox using correct terms, regardless if this means changing the prefix or the divider. 22.53.23 # I have to agree with linuxstb and Bagder on this, while the ISO kilo is definitely 1000, I've always knew that in storage a kilobyte is 1024 bytes. That's what I've been constantly encounter except storage manufacturers 22.53.56 # bluebrother: most drives have had that on their label for years now 22.54.12 # XavierGr: Why is it a problem to adjust that KiB *definitely* means 1024 bytes, rather than kB *probably* means 1024 bytes? 22.54.16 # Why cling to this uncertainty? 22.54.39 # XavierGr: only because it has been used that way doesn't make it more correct. And we have a correct way since several (!) years now, so what's the point in not following it? 22.55.29 # * petur thinks KiB, MiB,... sound silly 22.55.37 # I accept the logic of using KiB, I just don't like the way "KB" has been stolen from me by marketing people... 22.55.46 # And I agree with petur... 22.55.52 # rasher, bluebrother: well other than that some users might not be used to KiB, I don't find a reason not using it. Though it would be silly to use kilo as 1000 in this context 22.56.07 # AND it uses 50% more bytes to store! 22.56.16 # petur: rofl 22.56.38 # * bluebrother hands petur some bytes 22.56.58 # * petur wonders if they go with his current beer 22.57.08 # linuxstb: Stolen by marketing people? Would you listen to yourself! You've been using the *factually and objectively wrong* terms the entire time, and now you complain that you don't want to use actually correct terms because that's what marketing guys do? 22.57.13 # * rasher boggles 22.57.34 # XavierGr: using KiB is even an advantage IMO -- people won't expect it to be the same as the kB that's printed on their disc (ok, this would rather be GB ...) 22.57.53 # in the early days, KB *always* meant 1024 ! 22.57.53 # so yeah I wouldn't mind seeing KiB and keep the /1024, but it would be very confusing to change it to /1000 and keep KB 22.57.59 # rasher: No, I've always been taught in computing that the multiplier was 1024. Which makes sense, because that's the natural unit. 22.58.12 # petur: true, but we aren't in those early days anymore ;-) 22.58.25 # petur: That's not true either. 22.58.27 # * petur heads for the time machine 22.58.27 # everything was better in the past *snif* ;-) 22.59.03 # One of the relevant wikipedia articles references very early material using powers-of-ten prefixes 22.59.07 # damn storage manufacturers and their evil marketing tricks 22.59.21 # rasher: in computing? 22.59.27 # XavierGr: Certainly 22.59.41 # maybe we should use KB to mean 1000 and 0xKB to mean 1024 :D 22.59.52 # damn IEC creating binary prefixes! Those bastards, only striving to end this wonderful confusion! 23.00.11 # display the space as hex value? 23.00.12 # XavierGr: bitrates should (continue to?) use Kbit/s 23.00.30 # let's go octal 23.00.33 # * bertrik thinks kibi and mebi sound silly too and can't remember the last time he heard someone using it (except for discussions how silly it is) 23.01.02 # well, is 128KBit 128*10^3 bytes or 120*2^10? AFAIK it's 10^3 so this is correct. 23.01.02 # bertrik: you rarely need to be very precise in speech, so you can keep using your imprecise units if you like 23.02.06 # What does LANG_SCANNING_DISK "Scanning disk..." mean? Are we looking for something on the disk? 23.02.16 # rasher: So do you always write MiB etc? 23.02.29 # linuxstb: I try to, whenever it makes sense 23.02.52 # When doesn't it make sense? 23.02.55 # It's really very rarely an issue 23.03.25 # linuxstb: if you're giving the size of an mp3 for example, using MB doesn't matter, because you're being awfully imprecise anyway 23.03.26 # num1: this is shown when dircache and/or the database scanns the disc for new/changed files, i think 23.03.56 # num1: it's not a search, but a look through the disk to "visit" everything 23.04.30 # rasher: to "visit" everything? 23.04.37 Quit avis (Read error: 104 (Connection reset by peer)) 23.04.43 # num1: inspect all files 23.04.56 # on a sidenote, anyone tried to battery benchmark a gigabeast? 23.05.11 # okay, I'm gonna translate this one as "Checking the disk" 23.05.15 # mine got only 6 hours with rockbox, haven't tested the OF yet. 23.05.58 # num1: you can also modify it later if you discover it doesn't make sense when using Rockbox 23.06.22 # rasher: that's a good point, thanks 23.06.37 # actually, I don't mind really what convention rockbox uses, as long as we're consistent about it 23.07.02 # num1: A translation you're not completely sure about is better than no translation imho. Errors can be easily fixed 23.07.28 # I don't mind that either, but I want it to be correct -- which currently isn't the case. 23.07.53 Join sin613 [0] (n=pbarton@71-209-110-238.farg.qwest.net) 23.08.31 Quit OlivierBorowski (Remote closed the connection) 23.08.43 Join jgarvey [0] (n=jgarvey@cpe-098-026-069-229.nc.res.rr.com) 23.09.07 Quit ompaul (Client Quit) 23.10.52 *** Saving seen data "./dancer.seen" 23.14.00 # I wonder if it might be possible to get someone "proffesional" to do a human voicefile. It's "only" 600 strings. 23.15.24 # rasher: If someone spent the time to make a decent GUI for recording voice clips for rockbox it wouldn't be a bad idea. 23.15.57 # rasher: Could be nice for 3.0... But then what languages do we pay for? (presumably we would need to pay...) 23.16.57 # Well, the people who are doing translations already are presumably fluent in the language, and you don't have to pay for them to talk into a can 23.17.23 # num1: not everyone's got great voices 23.17.55 # Anyone know how many words are in the English .lang? 23.18.01 # I more or less abandoned my own attempt mostly because the result was quite awful (beyond the fact that you always think you sound funny) 23.18.26 # well, an "awful" voice is rather subjective. 23.18.29 # linuxstb: wc -w says 1174 23.18.51 # (on a text file with raw voicestrings) 23.19.10 # Hmm, then that would be £1174 pounds with the person I was thinking of... 23.19.34 # (someone who did some British English voicing for Asterisk - http://www.enicomms.com/cutglassivr/ ) 23.19.59 # linuxstb: that's rather a lot 23.20.30 Join tessarakt [0] (n=jens@e180070044.adsl.alicedsl.de) 23.20.52 # Hopefully that price (1 GBP per word) is just for smaller jobs... 23.21.29 # shouldn't it be rather easy to write a script that presents you a string, waits for a button, then records until next button press and use that recording? 23.21.41 # maybe adding a retry button too 23.22.15 # bluebrother: that's what I was thinking about 23.22.24 # I found that reading 10-15 clips in one recording, then cutting it up in audacity was quite fast 23.22.37 # It would be easy to script it with Asterisk.... 23.22.52 # even if some people had bad voices, if you made a gui really simple to use enough people would do the voiceovers that you could choose the best ones 23.23.09 # i just gave my old boss a call. he does a radio show and might be willing to contribute. hopefully get a call back in a few minutes 23.23.46 # make it a contest ;-) 23.24.28 # lol, winner gets 10% of our income, bty we don't have any :D 23.24.54 # We do, actually - donations 23.25.26 # could always see if a geeky celebrity to do it pro bono ;) 23.25.47 # Wil Wheaton or something 23.26.19 # maybe a geeky musician 23.26.29 # who already gives away music 23.27.03 # hasnt Robin Williams been seen around Linux conventions? 23.27.08 # Jonathan Coulton's given away a lot of his music, and his voice's pretty good 23.27.35 Quit desowin ("KVIrc 4.0.0 Insomnia http://www.kvirc.net/") 23.28.23 # Musicians are probably even more likely 23.28.28 # the good thing about musicians is that you've got voice talent 23.28.38 # And the equipment 23.28.52 # yeah 23.29.02 # and more than likely the know-how 23.29.31 # They don't even have to do editing. They could just record the entire list in one giant take, and I'm sure we could take care of the editing 23.30.54 # it would likely be better that way 23.31.16 # i've got equip and knowhow... dunno bout the voice though, but i could to a test later and see if anyone likes it 23.31.34 # nice voice mic? 23.31.39 # yeah 23.31.43 # ze: I've a script that produces a clean list of strings 23.31.45 Quit snoh (Remote closed the connection) 23.31.59 # what model? 23.32.01 # ze: FS#7650 23.32.07 # oktava mk219 23.32.11 # rasher: In one giant take? You want to sit in front of a mic for 1174 lines? that's almost 20 minutes stright! 23.32.43 # another possibility would be asking someone from a popular tech or recording podcast 23.32.44 # num1: well they could pause or do a couple of takes :) I just mean that they don't have to do the boring work of cutting it into tiny chunks 23.33.10 # There are plenty of good options, really 23.33.34 # podcast producer would be great since it would likely increase exposure for rockbox 23.34.16 # people from other countries might be able to locate podcast producers in their native toungue to serve as voice talent 23.34.32 # sin613: a good point 23.35.06 # it's a given they'll already have the equipment, the voice, and the know-how 23.35.24 # we'll never get that done before 3.0 unless we delay it 23.35.27 # if they get credit and more traffic, they'll likely do it gratis 23.35.50 # bertrik: Luckily, voicefiles can be added retro-actively 23.35.54 # There are people that do podcats with highly annoying voices :) 23.36.00 # podcasts too 23.36.17 # BigBambi: I'd wager it's most of them! 23.36.24 # :) 23.36.24 Join MaartenHD [0] (n=56572815@gateway/web/cgi-irc/labb.contactor.se/x-238dc4960abd925d) 23.36.57 # that's true, but there are many out there with good voices 23.37.24 # It'd be nice to have for e.g. English a couple with different accents too 23.37.27 # now we need to delay 3.0 ... ;-) 23.37.36 # e.g. UK, US, Aus 23.37.57 # the voice files weren't going to bundled with the release, were they? 23.37.59 Quit kugel ("ChatZilla 0.9.83 [Firefox 3.0.1/2008072820]") 23.38.14 Join kugel [0] (n=chatzill@unaffiliated/kugel) 23.38.25 # no, you download it separately. 23.38.26 # Hmm, just got a giant time out without noticing 23.38.41 # so will it matter if they're released at the same time? 23.38.45 # but there is a remaining issue: rbutil only installs the most recent voicefile, there is no selection. 23.38.55 # I already thought there was no talk here since 3 hours 23.39.03 # if only there were some way to get at the source and change that... j/k 23.39.04 # * petur posts another recscreen patch and nudges pixelma 23.39.06 # if we really want this, we should do some modifications to our voicebuilding script, so that we can easily regenerate the voicefiles from the source wavs.. 23.39.23 Join Imaginal [0] (n=travis@74-60-4-104.sea.clearwire-dns.net) 23.39.36 # doesn't the voice script already allow for a voice pool? 23.39.39 # domonoky: it's not hard to do with a simple wrapper script that calls the buildsystem - see FS7650 23.39.40 # bluebrother: Does rbutil still (I think the wx version did) have an option to choose a Rockbox release when installing? 23.39.50 # linuxstb: yes. 23.39.59 # When I click through database-> recently added, nothing shows up. How do I make this feature work? 23.40.10 # Imaginal: initialize the database. 23.40.28 # or update it if you already have initialized. Keep in mind that you need to reboot after initializing the database 23.41.24 # bluebrother: the database has been initialized. I can browse other options, and autoupdate is turned on. I frequently reset it 23.41.51 # reset what? The database? 23.41.57 # the ipod 23.41.58 # bluebrother: There's also the manual as well... 23.42.12 # linuxstb: true ... 23.42.28 # Imaginal: Why do you reset often? 23.42.35 # Imaginal: resetting (holding MENU+SELECT) is like pulling the plug on your PC. You should do a clean shutdown (hold PLAY for a couple of seconds) 23.43.42 # understood. I meant that I've been changing a lot of things with it that require resetting, so it has been reset since initializing and updating 23.44.17 # Imaginal: changing options in Rockbox *never* require a *reset*. They require a *reboot*, which means you need to shut down cleanly 23.44.20 # What requires resetting? 23.44.27 # otherwise you can expect problems. 23.44.30 # I can't think of anything tht needs a reset 23.44.43 # hmm. cool 23.44.48 # ]If you reset, rockbox doesn't get chance to write the status to disk 23.44.59 Quit jhulst (No route to host) 23.45.08 # resetting is almost *never* needed. 23.45.41 # * bluebrother sighs. Once you disclose the reset procedure users tend to use that all the time 23.45.50 # but recently added seems like something the involves the transfer time, and would be writter at the beginning of the transfer 23.46.12 Join safetydan [0] (n=evertond@rockbox/developer/safetydan) 23.46.13 # The disk isn't spun up on every track change, that would kill battery 23.46.57 # Most things are kept in memory then written when the disk is spun to read more music into memory. If you reset, the data held in memory never has chance to get written to disk (which happens as part of the shutdown procedure) 23.47.32 # Imaginal: It's hard to diagnose if there's a problem because you've been resetting your ipod. Try adding some files and doing a clean shutdown, and see if you still have problems. 23.47.54 # resetting the Ipod when not needed isn't a good idea anyway, regardless of the state of the currently running software 23.48.07 # * BigBambi just realised he didn't quite understand Imaginal's last sentence 23.48.08 # I understand. I'll give it a try and get back to you. Thanks again 23.49.04 Quit Imaginal ("Lost terminal") 23.49.20 Quit MaartenHD ("CGI:IRC (EOF)") 23.52.44 Quit tarbo ("leaving") 23.53.36 Join tarbo [0] (n=me@unaffiliated/tarbo) 23.56.21 # I believe "recently added" needs the "Gather Runtime Data" option to be set. Not sure if it is in the manual but I think I read it on the DataBase wiki page 23.56.49 # bleh, didn't see he already left 23.56.54 # * kugel agrees with bluebrother and rasher on the previous discussion on KB/KiB 23.57.12 # petur: looking 23.57.42 # pixelma: maybe wait a moment, I seem to have a drawing issue in the ondio sim 23.57.47 # pixelma: I added detection of wrong "" vs none on my language problems page 23.58.03 # rasher: nice :) 23.58.48 # I don't think I'm going to change the logic on the translation page, because it really is translators not being awake, and at least like this, the string is kept, even though it's in the wrong place 23.59.00 # Has anyone noticed the go-back-behavior of wps context menu items being very inconsistent?