--- Log for 17.03.106 Server: clarke.freenode.net Channel: #rockbox --- Nick: logbot_ Version: Dancer V4.16 Started: 11 days and 23 hours ago 00.00.42 Quit webmind_ (Connection reset by peer) 00.02.09 Quit bluey (Remote closed the connection) 00.06.01 Join webguest91 [0] (n=3e4f4094@labb.contactor.se) 00.06.40 Quit midkay (Read error: 104 (Connection reset by peer)) 00.06.49 Join midkay_ [0] (n=midkay@24.143.70.99) 00.07.16 Quit midkay_ (Read error: 104 (Connection reset by peer)) 00.08.11 # wow, this starbox is pretty cool, http://www.origami.cz/Bin/starbox.gif 00.08.31 # it's been a while since i did origame... took me like 15 minutes 00.09.15 # unasmking the timer2 interrupt just causes everything to hang 00.09.20 # so it's clear i'm not acking the correct interrupt 00.10.10 Join JdGordon [0] (n=JdGordon@c211-28-227-249.smelb1.vic.optusnet.com.au) 00.10.14 # * webguest91 lightly curses vim 00.13.25 # KE KE KE KE 00.13.28 # what did it do now? 00.13.30 # Having \n mean null isn't what I'd have expected. 00.13.33 # system.c is becmoing a bloody mess thanks to the ipods 00.13.40 # i added the interrupt handler in the wrong section 00.13.47 # um, it doesn't mean null 00.14.07 # heh 00.14.45 # Mikachu: in vim substitutions it does 00.14.52 Join DrumRBoy320 [0] (n=Drumrboy@ool-44c2019c.dyn.optonline.net) 00.14.57 # * petur feels he's not ready yet to analyze disassembly :( 00.14.59 Quit webguest91 ("CGI:IRC") 00.15.01 # hello everybody 00.15.03 # it doesn't... 00.15.04 # petur: what'cha looking for? 00.15.04 Join webguest91 [0] (n=3e4f4094@labb.contactor.se) 00.15.06 # Anyone have any suggestions about how often I should call rb->yield() from pacbox. Currently it's never called... 00.15.12 # linuxstb: nice 00.15.25 # linuxstb: it's a pretty empirical thing, just try it 00.15.32 # does RB for teh 5g iPod 60g support the full 64Mb ram yet? 00.15.33 # preglow: what iriver's doing with regards to usbotg 00.15.37 # probably a couple of times a frame 00.15.45 # DrumRBoy320: no 00.15.58 # thats abummer 00.16.00 # does it need to be that much if audio isn't playing at the time? 00.16.02 # like what is this: move.l #-0x20000001, d0 00.17.03 # I know you guys cant really give ETAs on things or features but is it hard work to get EQ working on Ipod5G along with smooth mp3 playback ? 00.17.38 # kclaf: doesn't eq and mp3 play together? 00.17.42 # Mikachu: :help sub-replace-special tells me "\n insert a ( in the file)" 00.17.42 # kclaf: it does on nano 00.17.49 Join Doomed_9 [0] (n=nnscript@ool-4353c24e.dyn.optonline.net) 00.18.01 # kclaf: What bitrate are your MP3s? 00.18.03 *** Saving seen data "./dancer.seen" 00.18.07 # 192 00.18.09 # or VBR 00.18.12 Nick Doomed_9 is now known as Doomed (n=nnscript@ool-4353c24e.dyn.optonline.net) 00.18.19 Nick Doomed is now known as Doomed9 (n=nnscript@ool-4353c24e.dyn.optonline.net) 00.18.22 # webguest91: oh you mean on the righthand side 00.18.23 # petur: move.l #efffffff, d0? 00.18.33 # webguest91: just press ctrl-v enter i think 00.18.42 # kclaf: Basically, the real problem is just that the mp3 codec needs to be optimized more for arm. 00.18.42 # hmmm is it that simple... 00.18.45 # petur: kclaf do you mean just when you are adjusting the eq? 00.18.51 # preglow : it plays together, but mp3 is not smooth at all 00.18.59 # Mikachu: Ah, yes.. multiline matching is a thing of the devil.. and yes, ctrl-v enter indeed. 00.18.59 # no preglow 00.19.04 # petur: i don't know, that's just from what i'm remembering about two's complement 00.19.04 # all the time 00.19.07 # kclaf: really? 00.19.12 # that's bad 00.19.27 # yea, sound cuts like every 10 seconds 00.19.28 # webguest91: i can agree that is a bit weird 00.19.34 # petur: god knows why objdump choses to format a constant like that 00.19.34 # i would put \0 as null if anything 00.20.07 # kclaf: well, no promises, but i'm going to work on optimising mp3 on ipod in a little while 00.20.15 # Mikachu: That's match though 00.20.21 # That is, backreference 00.20.25 # ah yeah 00.20.29 # i should perhaps try to use a less complicated theme, could that help ? 00.20.36 # im currently using brushed metal 00.20.38 # regexes always make my head hurt 00.20.43 # kclaf: A simple theme, and not having a backdrop image, would both help. 00.20.45 # kclaf: try going out to the menu 00.20.55 # ok, second 00.21.04 # (i'm assuming the wps isn't drawn underneath the menu) 00.24.25 # ok, playing from the menu makes it smooth 00.25.07 # i guess i have to use some other theme until mp3 decoding is optimized 00.28.14 # graphics is too bloody heavy on colour targets yet 00.28.18 # seriously needs to be fixed 00.29.18 # right, this usbotg will have to wait another week (holiday) 00.29.41 # see y'all in a week 00.29.43 Quit petur ("here today, gone tomorrow") 00.29.48 Join xmixahlx [0] (n=xmixahlx@64.122.111.98) 00.30.01 # and i have a fading backlight! 00.31.18 # on ipod 5g? 00.31.39 Nick Spida_ is now known as Spida (i=Spida@p508A3C73.dip0.t-ipconnect.de) 00.31.42 # yep 00.31.46 # that is, nano 00.31.49 # but 5g as well, probably 00.31.58 # I'm sure :) 00.32.01 Quit SereRokR ("XChat Aqua") 00.32.30 # it'll need quite a bit of restructuring before i can commit it 00.32.50 # the code was heavily h1x0 centric and duplicated code everywhere 00.33.23 # so the ipods have a variable backlight, or is it done with pwm? 00.33.45 # pwm 00.33.52 # it looks like 4g has some other way of doing it 00.34.05 # but the good news for me is that timer 2 works 00.34.11 # and that ipod now has a timer.c module 00.35.20 # so i'll just commit that part of it now and deal with the backlight later 00.35.49 # hmm, atm nothing that desperatly needs atimer comes to my mind, but as time goes by... 00.36.37 # grayscale lib on 4g grayscale and lower... 00.37.04 # hey, is the iriver 120 port still being worked on? 00.37.11 Join webmind [0] (i=webmind@feather.perl6.nl) 00.37.34 # Yes - no port is ever finished. 00.37.41 # preglow: I did some testing regarding the gpio on ipod 5g. The 8 in 0x6000d824 you have mentioned is insignificant. So using 0x6000d024 has the same effect (at least for me). 00.38.00 # herz42: not for me 00.38.14 # ohh 00.38.17 # herz42: that is, it depends 00.38.37 # all the 6000d824 writing i've seen uses 0x100 anded in with the value to be written 00.38.45 # if i remove that, then it doesn't work 00.38.55 # however, for 6000d024 it doesn't matter 00.39.18 # k 00.39.20 # so there is some difference 00.39.24 Nick scf is now known as Zzzcf (n=scf@141-113-207-82.ip.ukrtel.net) 00.39.47 # ouch 00.39.56 # backlight fading isn't as smooth as on h120 00.40.04 # probably due to the lower clock, then 00.40.15 # hmm, no, that shouldn't be so 00.40.21 # the switch interval is way higher 00.40.35 # hm...i wonder if it will ever play music when u play a rom 00.41.47 Quit [IDC]Dragon ("CGI:IRC (Ping timeout)") 00.42.06 # when we start using two cores, it should play music whatever you do 00.42.22 # preglow: Yeah, but I think Doomed9 was referring to the iRiver h1x0 port. 00.42.38 # yea 00.43.47 # In which case, it all depends on how well someone manages to optimize Rockboy and the codecs, I suppose. I really don't know how possible/impossible it is. 00.44.34 # right 00.44.42 # it's very possible 00.44.46 # but a boatload of work 00.45.00 # I mean, it'd also depend on the bitrate of your music, etc. 00.45.02 # Don't forget the audio buffer - Rockboy steals it AFAIK. 00.45.11 # So, rockboy is not very effecient? 00.45.17 # hell no 00.45.26 # rockboy and efficient probably shouldn't even be in the same sentence. 00.45.44 # Paul_The_Nerd: wanna test some sweet sweet backlight fading? it'll b0rk your settings 00.45.56 # preglow: I am *always* up for b0rked settings 00.46.03 # * JdGordon 2 00.46.33 # is there a clock in rockbox? 00.46.36 # http://www.pvv.org/~thomj/rockbox/rockbox.ipod 00.46.41 # Doomed9: depends on your player 00.46.44 # Doomed9: h1x0 has no clock 00.46.47 # damn 00.46.53 # h3x0 and ipods do 00.46.56 # plus some archoses 00.47.08 # booo!. do a h300 version 00.47.20 # of clock?? 00.47.31 # of backlight fading 00.47.39 # amiconn tells me it's not possible 00.47.44 # backlight is controlled indirectly 00.47.49 # * Paul_The_Nerd wonders why he just found a rockbox.iriver on his ipod 00.47.54 # lol 00.47.59 # Paul_The_Nerd: i think it's coming out of the closet 00.48.06 # preglow: i guess you already have a nano to test on 00.48.06 # rename it to .ipod and see what happens :p 00.48.16 # Mikachu: yeah 00.48.22 # preglow: Oooh... nice. It fades nice and smooth and everything. 00.48.28 # what i put up is a nano binary so... 00.48.37 # Paul_The_Nerd: gr8 00.48.55 # not h120 smooth, but it works as long as you don't use 10s or anything 00.48.59 # has anyone thought of allowing escape characters in filenames, so you could keep the free-form structure that files give you, but use all available characeters (like \ and / and ? and .) in names? 00.48.59 # try that, and you'll see it flicker 00.49.00 # s 00.49.26 # 10sec isn't completely smooth on h1x0 either 00.49.27 # nudel: fat doesn't allow ? 00.49.37 # amiconn: not completely, but this is a tad bit worse 00.49.46 # but you could have like [xx] where xx is the hex code and it woudl turn it into ? 00.49.49 Join midkay [0] (n=midkay@c-24-16-191-240.hsd1.wa.comcast.net) 00.49.59 # * amiconn says goodnight 00.50.02 # preglow: Hey, while you're doing backlight stuff, any idea why "Off" has the same result as "On?" 00.50.05 # nn amiconn 00.50.11 # Paul_The_Nerd: wha? 00.50.14 # hm...would it be easy to make a clock plugin? 00.50.21 # nudel: you mean the filename would be the actual string [2f] and rb would interpret it as /? 00.50.23 # preglow: Setting backlight to "Off" leaves it permanently on. 00.50.30 # Paul_The_Nerd: does it now 00.50.33 # Paul_The_Nerd: did it always do that? 00.50.34 # yeah Mikachu 00.50.38 # Doomed9: why not use the clock plugin? 00.50.43 # nudel: ah, that might be possible... 00.50.46 # lol im new 00.50.50 # preglow: I looked at it a little bit, found myself confused. I managed to make it so "Off" turns it permanently on, and "On" permanently off, but couldn't reverse it. 00.50.59 # preglow: AFAIK, it's always done it on iPod 00.51.07 # Paul_The_Nerd: right, then i'll see if i can discover why 00.51.23 # but not now 00.51.57 # Fair enough 00.52.01 # I set mine at 2 seconds anyway 00.52.03 # ;-) 00.52.48 # preglow: Just one more thought: If 10 sec isn't smooth, then probably something else fiddles with th eport during the fade 00.53.17 # preglow: can i have a diff? 00.53.24 # Mikachu: the diff is really, really ugly 00.53.24 # We had that way back on H1x0, where the fade started flickering like mad when something was drawn on the remot elcd 00.53.30 # amiconn: and contains other stuff 00.53.39 # amiconn: it's irregular in a really consistent way 00.54.03 # okay, i'll just wait then, i've fiddled with too many things for a binary to work with my plugins probably :) 00.54.05 # That lead me to implement the atomic port bit manipulation 00.54.07 # amiconn: and why should the other port pins interfere with the backlight one? 00.54.29 # preglow: Perhaps some place doesn't use the masked write address? 00.54.47 # amiconn: perhaps, but in this case i think everything is right 00.54.53 # You said this special '8' in the port address doesn't matter - perhaps it does? 00.55.06 # amiconn: have christi try it out on hers at devcon and have a look for yourself :) 00.55.15 # amiconn: i've kept it 00.55.36 # Yes, *you* have. Is there other code that accesses the same port? 00.55.46 # ahh, yeah 00.55.50 # but i don't think there are any writes 00.56.22 # :( 00.56.27 # i'll see, the backlight handling needs a lot of fixing before i'll commit it anyway 00.57.08 # amiconn: and besides, it's only noticable in the very darkest levels 00.57.24 # amiconn: looks more like pwm steps than port garbage to me 00.57.32 Join DJ_Dooms_Day [0] (n=DJDD@220-245-186-182.static.tpgi.com.au) 00.59.24 Quit moozooh (Read error: 131 (Connection reset by peer)) 01.02.33 Quit Paul_The_Nerd ("Leaving.") 01.02.36 Join moozooh [0] (n=moozooh@87.240.1.66) 01.02.45 Join damaki_ [0] (n=Chocolat@ALille-153-1-42-176.w83-198.abo.wanadoo.fr) 01.06.09 Quit ender` (Read error: 104 (Connection reset by peer)) 01.06.30 # hm...theres no wma support in H120? 01.07.03 # theres no wma support at all 01.07.08 # :( 01.07.33 Join XavierGr [0] (n=XavierGr@ppp90-adsl-2.ath.forthnet.gr) 01.07.44 # bins 01.07.48 # oops? 01.07.52 # there's been some work done on it 01.07.59 # but it's not anything approaching complete 01.08.51 # kk 01.08.56 # wtf??? ive got this line in a lyrics file for the track im listening to... "-- Solo duet --" WTF?? 01.09.40 # hm...i guess its hard to take a wma codec and convert it to w/e rockbox uses 01.10.08 # Doomed9: well, there is an opensource one in ffmpeg, but the problem is that it's a floating point implementation 01.10.33 # and since the embedded players tend to not have FPU's, anything using floating point math will be far far too slow 01.10.46 # kk 01.11.00 # do any of the targets have a fpu? 01.11.07 # if that happend, i wouldnt boot into the iriver firmware 01.11.09 # like i say, there's been some work on it, on and off, but it's going to require rewriting it to use fixed-point math, which is a fairly extreme undertaking 01.11.32 # it will probably get done eventually 01.11.38 # bah, tab alert 01.12.07 # preglow: did you or linuxstb ever get much done with the partial work that you picked up? 01.12.11 # i like all of these games 01.12.36 # ashridah: someone on the forums have started a newer effort 01.12.45 # * ashridah tends to just transcode wma to flac in the few rare instances that he runs into it 01.12.52 # ashridah: I spent an hour or so trying to understand the ASF format, but couldn't. 01.12.54 # preglow: ah, interesting. i haven't been reading those in a while 01.13.05 # ashridah wht do u use mp3? 01.13.06 Join bepe86 [0] (n=52c2c0c3@labb.contactor.se) 01.13.18 # http://forums.rockbox.org/index.php?topic=3005.0 01.13.46 # linuxstb: you think i should commit cpu boost support for 5g ipods? 01.13.56 # Doomed9: ogg mostly, some mp3, since there's no value in transcoding it, and i haven't got originals for some of them (got robbed 5 or so years back) 01.14.25 # preglow: I don't like to see the 4g fall behind, but yes, I think it's worth committing it. 01.14.37 # ogg the same size as wma ?? 01.14.55 # i have no idea, they're both lossy codecs 01.15.00 # linuxstb: k, i'll be glad to have it off my tree 01.15.14 Join elinenbe [0] (n=elinenbe@207-237-225-94.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com) 01.15.16 # yuea 01.15.22 # but i almost never use wma, so i have no idea what its quality/rate scale is 01.15.29 # it's crap 01.15.42 # and it's smelly 01.16.01 # preglow: you see, that's what i kinda assumed, plus it'd be a crime against nature if i started ripping to wma on a linux system :_) 01.16.02 # :) even 01.16.09 # My only wma file is actually quite listenable - it's a 48KHz 128kbps webcast. 01.16.29 # But every other wma file I've heard just gives me a headache... 01.16.41 # my only wma is a 64kbps file 01.16.43 # hmm.. 01.16.46 # and by god, it shows 01.16.54 # ogg or mp3? 01.18.09 # ogg's got a few minor advantages over mp3. firstly, it's got variable frame sizes, so you don't need a dodgy hack to get gapless playback (although said hacks often work pretty well for mp3, and LAME can insert extra info that helps too) 01.18.10 # it' 01.18.19 # s slightly better in the quality/bitrate department 01.18.33 # but not so much so that it's worth reencoding all your mp3s 01.18.38 # anyway to convert my wma to ogg....without losing anything 01.18.42 # (and of course, transcoding would be twice the loss) 01.18.46 Quit damaki__ (Read error: 110 (Connection timed out)) 01.18.50 # Doomed9: You can't... 01.18.58 # Doomed9: would be impossible, they're both lossy, use the original, lossless, source 01.19.06 # or wait for a working codec 01.19.13 # yeah 01.19.24 # i dont have a source for most of mine 01.19.35 # * Mikachu fiddles with absolute wheel for pacbox 01.19.44 # is wma-pro a different codec or just a different encoder? 01.20.59 Join TiMiD [0] (n=TiMiD[FD@asgard.valombre.net) 01.21.08 # 'morning 01.21.42 Quit mikearthur (Read error: 104 (Connection reset by peer)) 01.21.52 # o well 01.21.53 Join mikearthur [0] (i=mike@82-41-205-190.cable.ubr11.edin.blueyonder.co.uk) 01.22.05 # preglow, linuxstb: I'm trying to figure out the port pin assignments from the disassembled drivers. Found out how the background controller, the touch pad and the LCD CS are connected. 01.23.03 # The touch pad is read as a 13 bit value, with each bit representing one of the 13 touch sensors. 01.24.07 # pretty lowres compared to the clickwheel 01.25.00 # how does the clickwheel work? 01.25.08 Quit obo (Read error: 110 (Connection timed out)) 01.25.22 Join TCK [0] (n=tckocr@81-178-155-77.dsl.pipex.com) 01.25.28 # you just get a number from 0 to 95 01.25.33 Join sersport [0] (n=sergavr1@ts18-a120.Moscow.dial.rol.ru) 01.25.36 # describes position 01.25.55 # linuxstb: http://mikachu.ath.cx/patches/rockbox-pacbox_absolute_wheel.patch 01.26.02 # hi 01.26.13 # linuxstb: it's really playable with that i think 01.26.20 # I live in Russia 01.26.27 # ok 01.26.41 # i hear lots of people do 01.26.50 # where are you from? 01.26.54 # USA 01.26.55 # ok. Well, at least we can have UP and LEFT at the same time (nice for rockboy) 01.27.07 Quit ashridah ("Leaving") 01.27.07 # we're from lots of places 01.27.11 # city? 01.27.21 # Enschede 01.27.23 # :) 01.27.46 # Roda 01.28.03 # there's a small town in sweden called enskede :) 01.28.29 # ok 01.28.48 # I love ''The Europe'' 01.29.00 # hey, I'm just wondering if it's possible to build the h1xx manual using cygwin, I'm having a few problems with it... :S 01.29.28 # sersport: 'the' doesn't exist in russian, right? 01.30.13 # yes 01.30.29 # bepe86: the manual will be available for download on the rockbox website soon. 01.30.37 # "---! /var/lib/texmf/web2c/pdflatex.fmt was written by pdfetex (Fatal format file error; I'm stymied)" 01.30.47 # okey, thanks 01.31.23 # I got it to build at first, but got like 100 warnings about missing fonts, and the pdf file itself was blank except for images 01.32.09 # life in Sweden is good 01.32.14 # i bet 01.32.25 # in Norway too ;) 01.32.28 Join tvelocity [0] (n=tony@ipa71.14.tellas.gr) 01.32.36 # yes, i'm not exactly suffering 01.32.48 # too bad you get booze and snus cheaper than us :S 01.32.59 # woo, another norwegian 01.33.08 # snus sucks anyway 01.33.16 # :> 01.33.23 # heh 01.33.47 # I try to learn Norwegian language 01.34.02 # ah, cool, good luck :) 01.34.31 # linuxstb: i finished 2 levels with the 5 pacmen setting, that's a first 01.34.53 # mange takk 01.35.05 # bare hyggelig 01.35.32 # or as we write in nynorsk, berre hyggeleg ;) 01.35.58 # sersport: varf�r inte svenska? :) 01.36.15 # Mikachu: wha, you're swedish??? 01.36.21 # tada 01.36.27 # i thought you were american for some reason 01.36.29 # god knows 01.36.36 # maybe i was being too annoying 01.36.40 # hahaha 01.36.48 # or maybe i was acting the fool 01.37.06 # man, it's this late already :S better go to bed then, see you guys later 01.37.10 # i have been known to do that 01.37.13 # bepe86: see ya 01.37.17 # preglow: for the first couple of days i read your nick as preg-low 01.37.20 Quit bepe86 ("Zzz") 01.37.26 # Mikachu: some people do that 01.37.35 # and you can continue to do so for all i care :> 01.37.41 # i used to be just 'glow' one day aeons ago 01.37.46 # ah 01.37.47 # then i suddenly got this urge to prepend something 01.38.30 # heh 01.38.40 # Do you visited Moscow? 01.38.48 # now that you all have top secret host masks it's harder to see your nationality 01.39.16 # but i know you're norwegian and linus is swedish 01.39.21 # Mikachu: it helps on my sexy undercover agent image 01.39.30 # ah, of course 01.40.06 # I'm russian 01.40.28 # and the first russian you are as well 01.40.30 # sersport: i figured when you said you live in russia :) 01.40.47 # a ton of swedes, couple of norwegians, one finn and now a russian! 01.41.34 # I love hockey, and you? 01.42.01 # i don't like most sports 01.42.35 # preglow: about absolute wheel, someone mentioned an interface to toggle the events mode and the absolute mode.. is there anything more? 01.42.52 # Mikachu: i don't follow 01.43.00 # ahh, right 01.43.06 # no, i've heard nothing more of it 01.43.07 # heh, maybe i talked to linuxstb about it then 01.43.09 # but i think we need it 01.43.14 # you have the same color 01.43.18 # hahaha 01.43.24 # perhaps... 01.43.55 # i've added that it returns -1 when you're not touching the wheel, i haven't put that in the patch tracker yet 01.44.10 # i'm not exactly sure why it works 01.44.29 # since it's only set in the interrupt handler, and that is only called when you touch the wheel.. 01.44.36 Join linuxstb_ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net) 01.44.37 # it works even if you don't press another button 01.45.01 Quit linuxstb (Read error: 110 (Connection timed out)) 01.46.12 Nick linuxstb_ is now known as linuxstb (n=linuxstb@i-83-67-212-170.freedom2surf.net) 01.46.39 # man, wav playback sounds like a bag of shit on ipod 01.47.00 # why should it? it's the least compressed 01.49.39 # hm...what other open source wma codecs are there 01.49.40 Join imphaing_ [0] (n=a766a239@yossman.net) 01.50.26 # Doomed9: None. We're lucky to have one - Microsoft aren't exactly open with their audio format specifications. 01.50.46 # does ours not work well? 01.51.00 # it seems we don't have one 01.51.15 # (in rockbox) 01.51.15 # If by "ours" you mean "rockbox" - then Mikachu is right, we don't have one. 01.51.37 # linuxstb: has wav playback used to work on ipods? 01.51.43 # linuxstb: did you see my patch? it is maybe not so useful yet 01.51.52 # http://mikachu.ath.cx/patches/rockbox-pacbox_absolute_wheel.patch 01.52.03 # preglow: Yes, it worked fine - I used a wav file when I was implementing audio playback. 01.52.09 # i find it's a lot easier to play when you don't actually have to press any buttons 01.52.32 # then i've probably messed up wav.c, then... 01.52.57 Quit Febs ("CGI:IRC") 01.53.09 # M$ should be more open, i mean considering most of their stuff is stolen 01.53.21 Part sersport 01.53.22 Quit imphaing_ (Client Quit) 01.53.23 # such as their wav header things 01.53.40 Quit Kohlrabi (Nick collision from services.) 01.53.50 Join Kohlriba [0] (n=Kohlrabi@dslb-082-083-128-245.pools.arcor-ip.net) 01.55.44 # Mikachu: I think that could be useful. It would be nice if you could update your wheel_status() patch. 01.55.58 # i just have the -1 thing, but i'll update it 01.56.28 # linuxstb: can you think of anything that uses endian swapping on ipod? 01.56.32 # aiff doesn't 01.56.46 # think i'll just do a printf test... 01.56.48 # Mikachu: Could it be combined with button_status()? 01.56.56 Quit TCK (Read error: 110 (Connection timed out)) 01.57.02 # linuxstb: i tried that but it won't work 01.57.19 # like putting the values 0-0x5f in the top byte in the return value 01.57.28 # but all code in rockbox checks ==, not & for the button value 01.57.49 # Even when using button_status() ? 01.58.02 # i think 01.58.35 # i don't know the code 100% yet so i can't promise anything 01.58.36 # I've just grepped, and in apps/, button_status is only called once, and it is checked using & 01.58.55 # ah 01.59.02 # then maybe yes 01.59.19 # but i think the way i set the wheel value in the int handler is a bit ugly maybe 01.59.34 # but i'll leave that for later 02.00.01 # It's also used in four plugins - brickmania, rockboy, pacbox and wav2wv 02.00.06 Quit webguest91 ("CGI:IRC (EOF)") 02.00.15 # pacbox uses & too 02.00.27 # i think brickmania uses case though 02.00.28 # Yes, and I would guess the others do. 02.00.51 # That's easily fixed though. 02.00.53 # or maybe not! 02.01.01 # yeah it uses & 02.01.42 # But I'm not sure if it's the cleanest solution. Maybe a separate wheel_status() is better... 02.01.54 # flyspray really needs a button to search for a 'opened by' 02.02.10 # should be pretty easy to change later if you want 02.02.46 # maybe someone could apply this in the meantime, http://www.rockbox.org/bugs/task/4789 02.02.54 Quit mikearthur (Remote closed the connection) 02.03.13 Join mikearthur [0] (i=mike@82-41-205-190.cable.ubr11.edin.blueyonder.co.uk) 02.05.27 # Mikachu: I think the idea behind highscore.cfg is that all hi-scores for all games can be stored there. Brickmania was probably the first game to use it. 02.05.56 # hm, okay 02.06.13 # it only has these contents 02.06.16 # file version: 0 02.06.16 # highscore: 3502 02.06.19 # But maybe I am wrong... 02.06.56 # I thought I remembered there being a "hi-score" library, but Brickmania is just using the standard configfile library. 02.07.29 # i'm not sure i like all those state files being in rocks/ either 02.08.02 # No, I was wondering about that as well. 02.08.21 # % ls *~*.rock 02.08.21 # bejeweled.save brickmania.score pacbox.cfg sokoban.levels 02.08.21 # bejeweled.score oscilloscope.cfg snake2.levels 02.08.29 # The configfile library puts them there. 02.08.45 # int new_wheel_value = ((status << 9) >> 25) & 0xff; 02.09.01 # is that different from status >> 16 ? 02.09.05 # yes 02.09.10 # the top bits are shifted away first 02.09.26 # aren't they masked away by &0xff anyway? 02.09.32 # yes, yes they are 02.09.32 # and that's doing sign extension 02.09.41 # herz42: no sign in that one 02.09.49 # Mikachu: god knows, it's ripped straight from ipl 02.09.52 # * linuxstb has a feeling of deja-vu 02.09.54 # okay 02.10.03 # wasn't the value +- 48 or so? 02.10.08 # mno 02.10.09 # i'm not going to touch it but it looked confusing 02.10.11 # it's 0..95 02.10.11 # herz42: 0-95 02.10.40 # Mikachu: i'm not going to beat you up if you discover redundancies 02.11.17 #