--- Log for 04.01.106 Server: kornbluth.freenode.net Channel: #rockbox --- Nick: logbot_ Version: Dancer V4.16 Started: 9 days and 22 hours ago 00.02.39 # do codecs have most of the IRAM to themselves or do they share a lot of it with rockbox itself? 00.03.02 # a codec or plugin gets up to 48k of iram 00.06.15 Quit matsl (Remote closed the connection) 00.07.53 # lostlogic: the stack of the codec and plugin thread are in rocbox kernel iram ? 00.08.11 # yes 00.08.36 Quit akaidiot (Connection timed out) 00.08.55 # Bagder: what do you think needs to be done with the RTC in WPS patch to get it in? 00.09.21 # I don't know, I haven't checked it 00.09.35 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 00.10.14 # oh, my bad, thought I'd talked to you about it before. 00.10.37 # might've been Bger 00.11.06 # could be... shrug. 00.11.29 Quit akaidiota (Connection timed out) 00.13.19 # Bagder: you can force arrays to both irams ? 00.15.46 Join akaidiota [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 00.25.07 Quit mirak (Remote closed the connection) 00.27.54 Quit JoeBorn (Read error: 110 (Connection timed out)) 00.31.07 Join RotAtoR [0] (n=e@12-210-82-91.client.insightBB.com) 00.31.50 Quit akaidiot (Read error: 110 (Connection timed out)) 00.32.59 Join ashridah [0] (i=ashridah@220-253-120-56.VIC.netspace.net.au) 00.37.16 Quit akaidiota (Read error: 110 (Connection timed out)) 00.46.16 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 00.50.26 # wow..got the sim working..awesome...you guys are great 00.58.27 Join akaidiota [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 01.11.24 Quit ender` (" How many Unix system vendors does it take to change a light bulb? None.All of the light bulbs you have are'standard variants) 01.14.38 Quit Kohlrabi (Nick collision from services.) 01.14.44 Join Kohlrabi [0] (n=Kohlrabi@dslb-082-083-129-122.pools.arcor-ip.net) 01.14.45 Quit akaidiot (Connection timed out) 01.15.07 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 01.19.14 # let's make sokoban preload all of it's levels into IRAM instead of getting them from disk every time it needs them. 01.19.14 Quit akaidiota (Connection timed out) 01.24.20 Quit Moos ("Happy new year @ all") 01.31.26 Quit muesli__ ("ich will Khe!!!") 01.41.10 Quit kro (Nick collision from services.) 01.41.10 Join kro_ [0] (i=ork@kro.hn.org) 01.42.48 # hmm, is the function 'strtok' not available to plugins? 01.42.54 # it seems to work in the sim if I include string.h 01.43.00 # but I only get an 'undefined reference' when compiling for the target 01.45.08 # there is no stdlib for rockbox 01.45.18 # if a plugin needs a function it must be in the plugin API or defined in the plugin itself 01.45.50 # but strtok is in rockbox/firmware/common/strtok.c 01.46.49 # so it seems to be written for rockbox 01.48.03 # and it appears to have a prototype in rockbox/firware/include/string.h 01.50.14 *** Saving seen data "./dancer.seen" 01.50.45 # so maybe it need to be exported to the plugin api if I want to use it? 01.53.56 Quit saa[b_r]ider (Read error: 110 (Connection timed out)) 02.04.57 # right, it would need to be in the plugin API if the plugins need to access it. 02.09.22 # bah, thanks 02.13.59 Quit akaidiot (Read error: 110 (Connection timed out)) 02.15.19 Quit Sacro () 02.15.41 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822) 02.21.03 # heh, copy & paste from strtok.c works... 02.27.20 Join ataxy [0] (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 02.31.49 Quit ataxy ("Chatzilla 0.9.69.1 [Firefox 1.5/2005111116]") 02.34.12 # grr 02.34.50 # plugin and codec IRAM are in the same place, so a plugin that uses IRAM can't be used when music is playing? that's sad. 02.39.25 Quit Thus0 ("Leaving") 02.55.09 Quit Kohlrabi (Read error: 104 (Connection reset by peer)) 02.59.47 # . 03.02.38 # How much RAM are plugins allowed to statically allocate? 03.04.53 Quit DreamTactix291 (Read error: 110 (Connection timed out)) 03.05.27 # IIRC it's 32KB for archos, 768KB for others (probably changing to 512KB, i think i heard) 03.05.36 # unless you want to grab the codec buffer 03.06.36 # 32kB is very limiting. 03.07.26 # sokoban has 36k of levels 03.08.40 # yuk 03.09.28 # maybe you could have it load the levels in two separate chunks? 03.09.39 # yeah, much more complicated that way, and I'm lazy ;) 03.09.49 # hehe :) 03.10.07 # probably just make it load 10 levels at a time or something. 03.15.28 Quit Maxime (Read error: 104 (Connection reset by peer)) 03.16.52 Join NibbIer [0] (n=sven@port-212-202-193-50.dynamic.qsc.de) 03.16.52 Quit saratoga ("CGI:IRC (EOF)") 03.18.19 Join Maxime [0] (n=flemmard@fbx.flemmard.net) 03.19.50 Join Jungti1234 [0] (n=jungti12@58.77.81.144) 03.26.48 Join DJDD_ [0] (n=DJDD@220-245-186-182.static.tpgi.com.au) 03.27.04 Quit Nibbler (Read error: 110 (Connection timed out)) 03.27.27 # Hiya 03.27.51 # hi 03.28.41 # I haven't been around for awhile, whats news with a iRiver rockbox? Any leaps and bounds made? 03.31.33 Join saa[b_r]ider [0] (n=saab_rid@221.223.98.68) 03.31.59 # hi saab 03.35.35 Join UmInAsHoE [0] (n=UmI@81-179-201-99.dsl.pipex.com) 03.36.07 # hello all 03.36.15 Part UmInAsHoE 03.37.02 Join UmInAsHoE [0] (n=UmI@81-179-201-99.dsl.pipex.com) 03.37.34 # back 03.37.36 # i am curious of what are the advantages of RockBox over the original firmware? 03.39.59 # ok forget that reading the the FAQ 03.40.05 # better reverse the question 03.43.29 # anyone here have it installed on their iriver h3xx? 03.44.13 # course 03.44.32 # do you find it alot better than the original firmware? 03.46.26 # While that seems like asking if I prefer mice or cdrws, I do significantly prefer rockbox. 03.46.59 # ok you lost me righ there? why are they so different? 03.47.05 # right* 03.47.25 Join mymomthelush [0] (n=lush@69-172-214-124.ontrca.adelphia.net) 03.47.37 # they both play music is how they are similar. Most everything else about using them is different. 03.47.52 # ah ok 03.49.22 Quit mymomthelush (Read error: 104 (Connection reset by peer)) 03.50.15 *** Saving seen data "./dancer.seen" 03.53.54 Join saratoga [0] (n=80c4c198@labb.contactor.se) 03.56.14 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 04.03.24 Quit actionshrimp ("a bird in the bush is worth two in your house") 04.03.25 Join bagawk [0] (n=lee@unaffiliated/bagawk) 04.07.34 # is there some sort of tutorial or idiots guide to rockbox programming? 04.10.51 # there is the code 04.11.30 # actually i dont know ;-) 04.11.32 # sorreh 04.14.50 # um 04.14.58 Join hd [0] (i=hd@gate-hannes-tdsl.imos.net) 04.17.36 Quit goa (Read error: 110 (Connection timed out)) 04.17.36 Nick hd is now known as goa (i=hd@gate-hannes-tdsl.imos.net) 04.22.34 Quit akaidiot (Read error: 110 (Connection timed out)) 04.37.24 # can rockbox use a iriver database file? 04.38.07 # patch 1396515 submitted to cache sokoban levels. Caches them _all_ (up to 2400) on high memory targets, and batches of 76 on the low memory targets, I think. 04.38.16 # (there was a previous patch out there, but it sucked. 04.39.13 # ty i think 04.39.28 # I dont' think rockbox uses the iRiver DB thing 04.39.34 # it has it's own tagcaching feature 04.39.40 # lol ok 04.42.58 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 04.49.49 Quit UmInAsHoE ("Yes I Quit. No You May Not Follow Me.") 04.56.43 Quit Jungti1234 () 05.03.30 Join Rob2222_ [0] (n=Miranda@ACB27B7B.ipt.aol.com) 05.09.50 Quit Rob2222 (Read error: 104 (Connection reset by peer)) 05.19.10 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au) 05.19.30 Join webguest14 [0] (n=8ddf56dc@labb.contactor.se) 05.22.50 # ??? 05.22.57 Quit webguest14 (Client Quit) 05.30.48 Quit DreamTactix291 (Read error: 110 (Connection timed out)) 05.33.01 Quit ashridah ("Leaving") 05.50.19 *** Saving seen data "./dancer.seen" 06.05.10 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 06.30.34 Quit darkless (Read error: 104 (Connection reset by peer)) 06.30.54 Join darkless [0] (n=darkless@62.79.44.48.adsl.vby.tiscali.dk) 06.32.13 Quit saratoga ("CGI:IRC (EOF)") 07.00.06 Quit paugh ("Leaving") 07.10.46 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 07.11.36 Join Paul_The_Nerd [0] (n=Paul_The@cpe-66-68-93-2.austin.res.rr.com) 07.16.00 Quit akaidiot (Read error: 110 (Connection timed out)) 07.35.48 # morning :) 07.36.19 # G'mornin. 07.36.59 # happy new year, Paul_The_Nerd ;) 07.41.46 # You too. 07.48.52 # 10q 07.50.23 *** Saving seen data "./dancer.seen" 07.54.16 Quit RotAtoR ("zzzzzzzz") 07.59.03 Join Jungti1234 [0] (n=jungti12@58.77.81.144) 07.59.37 # hey 08.00.48 # what's up in America? 08.01.36 # What happen is in America? 08.02.01 # The usual 08.04.03 Join joshn_454 [0] (n=kvirc@ool-182d4545.dyn.optonline.net) 08.05.22 Join Jungti1234_ [0] (n=3a4d5190@labb.contactor.se) 08.06.31 Join mirak [0] (n=mirak@AAubervilliers-152-1-54-189.w82-121.abo.wanadoo.fr) 08.14.02 Join ender` [0] (i=ychat@84.52.165.220) 08.20.59 Quit Jungti1234 (Read error: 110 (Connection timed out)) 08.23.32 # morning 08.24.33 # morning, amiconn 08.25.03 # Mornin' Amiconn 08.25.42 # lostlogic: There was an idea for the sokoban level cache some time ago: cache as many levels as fit into the remaining plugin ram, calculated dynamically 08.26.16 # there's a patch around for that 08.26.42 # [04:37:42] patch 1396515 submitted to cache sokoban levels. Caches them _all_ (up to 2400) on high memory targets, and batches of 76 on the low memory targets, I think. 08.26.49 # [04:37:52] (there was a previous patch out there, but it sucked. 08.27.17 # The 'batches of 76' seems to be a static amount 08.27.33 # hmm 08.30.01 # Speaking of ram... do flash based players usually use Ram to cache audio data? Like, I understand it so that the HD isn't constantly spinning, and doesn't have to spin up often. 08.32.36 Quit DreamTactix291 (Read error: 110 (Connection timed out)) 08.35.05 # Paul_The_Nerd: The one I know does (Ondio), and I think most of them do. The flash memory behaves as a mass storage device, not as directly addressable memory 08.35.49 # Aaah 08.35.53 # Of course the amount of buffer RAM is practically irrelevant for battery runtime 08.36.01 # That's more or less what I was asking about 08.36.17 # Like, the Nano has the same amount of Ram as other iPods, for example, and I was curious why that money was spent. 08.36.29 # At least I *think* it does... 08.36.38 # Even a very small buffer is sufficient because it doesn't matter whether the music is read in many small or few big chunks 08.37.15 # Having more RAM has advantages for other applications though, like an image viewer 08.38.41 Join olisto [0] (n=oliver@ppp-39-19.grapevine.net.au) 08.38.53 # Yeah 08.38.56 # I suppose that's true. 08.39.09 # I wasn't thinking in terms of "things it does other than playing music and running its OS" 08.40.13 Nick Jungti1234_ is now known as Jungti1234 (n=3a4d5190@labb.contactor.se) 08.40.18 # Keeping the audio buffer design has another advantage - it makes the system more similar to HD based players. Less engineering required 08.40.25 # Yeah 08.41.27 # That was actually my main interest in the Nano. To me, it was essentially an HD based player that used flash instead. I'm not generally a fan of flash players, but I wanted one. I also wanted something that was likely to be replaceable (Who knows when iRiver or someone else will stop manufacturing something, but Apple likes its monemakers) 08.42.50 # I thought that sudoku had a patch relating to caching levels go in sometime quite a while ago, actually 08.42.53 # Didn't it? 08.46.39 Quit Jungti1234 ("CGI:IRC (EOF)") 08.50.16 # http://svn.neurostechnology.com <= first Neuros code being available 08.51.02 # neuros equals blaah 08.51.13 # ;-) 08.51.13 # iriver equalas rockbox equals rocks 08.52.44 # Did you see the iRiver gaming device thingy? 08.55.17 # nope 08.55.43 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 08.56.26 # http://www.misticriver.net/showthread.php?t=34585 Probably named the "G10" 08.57.00 # neat beast 08.58.01 # iRiver has a knack for using names it shouldn't 08.58.01 # lol 08.58.21 # G# = logitech stuff 08.58.46 # * Rick has a Logitech G7 and G15 :) 08.58.57 # and G4 etc is PowerPC names used in macs 08.59.04 # hehe 08.59.23 # I wish I knew the specs. 08.59.34 # why is there mario kart on the screen? 08.59.35 # emulator? 08.59.41 # I was *thinking* about getting a GP2X. I want some sort of game player that's either hackable or open source. 08.59.49 # That's not Mario Kart. 08.59.55 # That's some other Mascot Kart game. 09.00.01 # ah 09.02.46 # I was thinking about a PSP, actually. With the right firmware, it seems like it'd be fun. 09.03.07 Join Zagor [0] (n=bjst@194-237-150-170.customer.telia.com) 09.03.32 # And the G10 probably won't hit stateside anyway 09.03.44 # Designed for Wireless Broadband gaming. 09.04.02 # ah 09.04.06 Join Jungti1234 [0] (n=jungti12@58.77.81.144) 09.04.07 # and yeah, I've got a PSP 09.04.16 # but you're screwed if you have firmware 2.01 or higher right now 09.05.11 Join _FireFly_ [0] (n=FireFly@pd95b7c08.dip0.t-ipconnect.de) 09.05.20 # Yeah 09.05.36 Join ender1 [0] (i=ychat@84.52.165.220) 09.05.47 # I'd pick one up used, so that I can make sure it's got older firmware, *and* I can try it in advance to make sure there aren't too many (or preferably any) dead pixels 09.06.48 # * Rick nods 09.06.52 # I've got one and a bunch of games 09.06.54 # 2.01 though :( 09.06.58 # so badly want to homebrew too 09.07.07 # thinking of picking up another one just for it 09.07.21 # I don't really want *any* of the PSP games, 'cept maybe Lumines 09.07.35 # ah 09.07.36 # I'm more interested in it as a homebrew/emulation device 09.07.38 # mercury is great 09.07.43 # you should pick that up too :P 09.07.46 # That's why I'm thinking about the GP2X as well 09.07.51 # ah 09.08.02 # the PSP is quite similar in specs as the upcoming Neuros device 09.08.10 # although using a somewhat older TI chip 09.08.17 # PMP 09.08.35 # oh 09.08.48 # * Bagder gets confused on the abbrevation front 09.08.56 # hehe :) 09.09.05 # Hehehe 09.09.14 # Too many names. 09.13.21 Quit ender` (Read error: 110 (Connection timed out)) 09.21.06 Quit akaidiot (Success) 09.34.56 # Misticriver's admin prevented my reply. 09.39.18 Join matsl [0] (n=matsl@1-1-4-2a.mal.sth.bostream.se) 09.40.18 Join ST [0] (i=STx@c211-30-160-92.blktn3.nsw.optusnet.com.au) 09.42.23 # hey do u guys know why the non-lcd remote patch hasnt been committed yet? 09.48.07 # because we need a way to use both remotes, not just one 09.48.15 # and for that we need a way to detect which remote is in use 09.48.25 # right. 09.48.36 # preglow that's not the only one problem ... 09.49.36 # http://portablevideo.engadget.com/2005/12/11/iubi-pmp2010-100gb-pmp/ <= wow 09.50.04 # heh 09.50.14 Join DangerousDan [0] (n=Miranda@newtpulsifer.campus.luth.se) 09.50.27 *** Saving seen data "./dancer.seen" 09.52.25 # http://www.iubi.co.kr/iUbiIntro-1/PMP2010_Overview.aspx 09.52.49 # <_FireFly_> wow 100GB 09.53.34 # yep :) 09.53.44 # VTR function! 09.53.51 # VTR ? 09.54.09 # yes 09.55.02 # video tape recording 09.55.17 # does harddisk really count as "tape"? :-) 09.55.23 # haha 09.55.29 # hdd 09.56.49 # it is called VCR in english normally 09.57.04 # ah.. 09.59.25 # http://www.iubi.co.kr/iUbiIntro-1/PMP2010_Feature.aspx 09.59.45 # ah? -_-; 09.59.55 # playtime : 3h 10.00.54 # very korean page 10.01.07 # yes :) 10.01.13 # haha Bagder 10.01.29 # Let's learn Korean! 10.02.03 # :P 10.03.39 # hahaha.. 10.08.11 # http://imgnews.naver.com/image/130/2005/11/24/5772_i_river320.jpg 10.09.33 # Good-looking face 10.09.41 Join amiconn_ [0] (n=jens@p54BD7964.dip.t-dialin.net) 10.10.29 # ? 10.10.57 # He is iriver president. 10.11.48 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 10.14.13 Quit ST () 10.16.37 # Rick: btw, PSP 2.01 is broken too, using the GTA save game hack. 10.17.28 # someone crack the 360! 10.17.31 # http://www.cdpkorea.com/zboard4/data/freeboard/cdpkorea-1136293764-1.jpg cowon iaudio6 10.17.48 # preglow: you mean the 360 portable? ;) 10.22.26 Quit Paul_The_Nerd ("Leaving.") 10.26.14 # Morning all 10.26.29 # more coffee! 10.26.36 # hi markun 10.26.40 Quit amiconn (Read error: 110 (Connection timed out)) 10.26.40 Nick amiconn_ is now known as amiconn (n=jens@p54BD7964.dip.t-dialin.net) 10.31.43 # Bagder: I did a little more disassembling http://130.89.160.166/Gigabeat/from.asm.bz2 10.32.22 Quit DreamTactix291 (Read error: 110 (Connection timed out)) 10.34.40 # looks like your mime.types need an update 10.35.42 # ok, I'll check. 10.36.05 # ah, yes :) 10.38.28 # preglow: whould it be application/x-bzip? 10.42.35 # x-bzip2, afaik 10.43.54 # Hm, didn't solve a thing. 10.44.50 # *shrug* 10.44.57 # been a little while since i've dealt with that 10.45.02 # coffee break! 10.45.13 # heh 10.47.56 # Coffee is bitter. 10.48.07 # huh, David Bryant is 2nd in credits' list of monkey's audio 10.53.58 # :-) 10.54.25 # how well does monkeys audio fare these days? 10.54.32 # compression wise, that is 10.54.48 # i don't know, really 10.55.06 # maybe google search ? :) 10.55.46 # hello jungti, markun, Bger 10.56.01 # good day saab 10.56.06 # hello everyone :) catching up with the interesting links! 10.56.08 # there is a comparison page at http://flac.sf.net/comparison.html 10.56.16 # hello, saa[b_r]ider 10.56.16 # hi saab :) 10.56.29 # but i don't know how recent is it 10.56.34 # (about the comparison) 10.57.26 # huh, the compared version of wavpack is 3.97, and it's not recent at all 11.01.02 # preglow : algorithm-wise, is there any probability of wavpack to become the "least cpu use" winner again ? 11.01.08 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 11.01.21 # Bger: i don't know, i hope so 11.01.50 # more can be done with wavpack, like putting more code in iram 11.02.52 # http://www.hydrogenaudio.org/forums/lofiversion/index.php/t8024.html 11.03.40 # Because of the way FLAC is designed, if a bit error happens somewhere in the FLAC file, only that block is affected. In most other lossless formats (including current versions of MAC and WavPack), a single bit error can corrupt the entire file beyond that point. <= is this true ? 11.03.57 # no 11.04.04 # wavpack handles frame errors very nicely 11.04.12 # it just mutes the current frame and moves on 11.04.14 # afaik 11.05.04 # this is said by "bryant" @hydrogenaudio.org. .. 11.05.21 # but huh, this thread is ooold 11.05.24 # just saw this 11.06.28 Join Kohlrabi [0] (n=Kohlrabi@dslb-082-083-129-122.pools.arcor-ip.net) 11.06.32 # in any case, i'm pretty sure it's fixed 11.11.00 # http://members.home.nl/w.speek/comparison.htm <= dated 2005.02.07 11.14.55 # so it does slightly better than wavpack 11.15.54 # preglow: at a huge complexity expense 11.15.59 # yeah, i know 11.16.07 # and the license is anal-retentive 11.16.09 # but with extra high mode ... which goes up to 5% CPU use while decoding on celeron @ 866 11.17.46 # wow optimFROG decoding is only 1.5 realtime on athlon @ 800MHz .. 11.17.47 # anyway the license doesn't matter just reimplement it and it should be fine 11.19.35 # let me refrase, the license might be valid for the one who is reimplementing it 11.19.40 # well... 11.19.54 # i'd rather just use something with a nice license in the first place 11.20.03 # yeah 11.21.49 # maybe ffmpeg people ? ;) 11.22.43 Quit akaidiot (Read error: 110 (Connection timed out)) 11.24.08 # we need more of them 11.25.26 Join Pi [0] (n=Pi@80-44-103-155.dynamic.dsl.as9105.com) 11.27.43 # is wavpack superior compared to flac? 11.28.03 # yes 11.28.08 # how well does rockbox deal with wavpack? 11.28.19 # according to my tests, it's a faster and more efficient encoder 11.28.21 # maybe I'll start using it then 11.29.17 # i've never seen wavpack produce a bigger file than flac, at least 11.29.32 # about the only good thing i can say about flac is that the rockbox decoder is veyr fast 11.29.36 # very... 11.30.18 # flac has more portable hardware support 11.32.16 # but for me there is no seeking in flac, and I dont want to hassle with it, so maybe I just start ripping to wavpack instead 11.32.29 # <_FireFly_> preglow: on wavpack.com there is a tiny-version of an example decoder 11.32.40 # <_FireFly_> novimon: seektables 11.32.44 # _FireFly_: we're using an updated version of that 11.32.55 # <_FireFly_> ah ok 11.32.59 # <_FireFly_> novimon: http://forums.rockbox.org/index.php?topic=2284.0 11.33.16 # _FireFly_: david bryant himself is developing it for us 11.33.18 # whats the easiest way of applying replaygain, btw 11.34.12 # <_FireFly_> preglow: :) 11.34.54 # novimon: well, i think using foobar2k is pretty easy 11.36.33 # do unsigned variables take up 8 bits? 11.36.45 # Pi: unsigned chars do 11.37.34 # what about plan unsigned (what is used to store colors) 11.37.51 # plain "unsigned" shouldn't be used imho as it looks weird 11.37.57 # but that defaults to ints 11.38.07 # and ints are 32 bits generally 11.38.14 # thanks 11.39.14 # does rockbox suport wavpack hybrid format? 11.39.21 # not yet 11.39.53 # Bagder: but plain 'long' isn't weird? 11.40.14 # long is a variable type, "unsigned" is more of an attribute in my eyes 11.40.24 # i think 'long' is the same 11.40.34 # the lossy .wv @ 256kbps is rough equal to 160kbps mp3 according to www.wavpack.com .. and 384 lossy .wv is fully transp. 11.40.35 # originally, yes 11.40.39 # <_FireFly_> novimon: but it plays fine the *.wv files 11.40.41 # but I don't view it as such 11.40.46 # * amiconn always just uses 'unsigned' for 'unsigned int' 11.40.51 # amiconn: me too 11.40.56 # It's shorter that way 11.40.56 # hardly anyone uses "long int" these days 11.41.06 # Bger: poor comparison, the noise profile is completely different 11.41.22 # preglow i said *rough* 11.41.25 # well, typedef it to 'u' and it'll get even shorter :-P 11.41.39 Quit matsl (Remote closed the connection) 11.41.43 # preglow: Do you know what type the wavpack losssy encoding is? Is it comparable to mp1/mp2? 11.41.49 # anyway, afk 11.43.25 Join muesli- [0] (n=muesli_t@141.71.4.201) 11.45.02 # amiconn: it's most likely not a transform domain codec, eg all processing is done in the timedomain 11.45.23 Join Pi_ [0] (n=Pi@80-44-103-155.dynamic.dsl.as9105.com) 11.50.29 *** Saving seen data "./dancer.seen" 11.51.10 Join LinusN [0] (n=linus@labb.contactor.se) 11.55.03 Join zamzon [0] (n=52875848@labb.contactor.se) 11.57.58 # amiconn: not at all 11.58.08 # can somone help? http://www.misticriver.net/showthread.php?t=34414&page=3 11.58.10 # I've created a rough document on how to program a plugin for rockbox (to help with beginners), and I'm going to add it to the wiki, but I have no idea what I should call the page. 11.58.14 # amiconn: it's a purely noise shaping time domain quantising thing, i believe 11.58.40 # amiconn: layer1/2 uses subband coding, which is inherently lossy, and so can't be used 11.58.40 # aliask: HowtoWritePlugins ? 11.59.04 # Sounds good to me. 11.59.45 Quit Pi (Read error: 110 (Connection timed out)) 12.00.03 # Hi! I'd like to get some advice in implementing some rockbox enhancements. I've just enabled the line-in while playing mp3 by adding two settings to the sound menu. (line-in volume and decoder volume). Now there is another parameter to set: ADC gain for the line in. Either I'll make another new sound setting or try to get the global setting for line-in gain (for recording) What do you think? 12.01.58 # Pi_: there are two storage spaces 12.02.47 # but woulent foreground and background be better stored in the firs space? 12.03.26 # Sorry, forgot to mention. It's for recorders and I want to be able play DJ a little or have my navi talking to me, while listening music.. :-) 12.04.04 Quit Jungti1234 () 12.05.48 # like archos recorders? 12.06.00 # yup. 12.06.02 # those settings would probably work iriver too, in that case 12.06.18 # have to check the registers 12.07.47 # Pi_: it doesn't matter, since both are stored on disk on the color platforms 12.08.32 # ok i will put it in the other place and try, thanks 12.12.30 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 12.12.37 # thanks works exactly as I want it to do 12.13.53 # i'd like to make the status bar better looking on the h100 and h300 12.14.19 # larger, multicoloured icons 12.14.55 # but then i realize that if i do, so many rockbox oldtimers will shout and complain about the larger status bar 12.15.06 # haha 12.15.08 # then i'll be forced to make it an option 12.15.14 # don't they all use archos units anyway? :) 12.15.17 # and the code gets larger and larger... 12.15.25 # so i dn't bother 12.15.34 # happens far too often nowadays 12.15.37 # hahah 12.15.43 # grumpy old man 12.15.59 # indeed 12.16.20 # * preglow calls for custom widget sets! 12.16.41 # read the "HTML UI" thread on misticriver and weap 12.16.58 Quit muesli- ("ich will Khe!!!") 12.17.38 # ahahahah 12.17.43 # the topic alone will do for me 12.17.51 # that 8-pixal status bar will be pretty small on the gigabeat :-) 12.17.57 # yup 12.18.09 # but there's got to be some way of customising it to each unit without increasing code size? 12.18.25 # 8 pixels font on 240 pixels hight... 12.18.29 # 30 lines! 12.18.29 # 320, no? 12.18.39 # true 12.18.39 # yes, 320 height! 12.18.47 # so 40 lines 12.18.51 # damn 12.19.08 # a wps-like status bar config would be cool 12.19.15 # we'll need it sooner or later 12.19.27 # lets hope some gui equivalent of slasheri comes along 12.19.41 # * LinusN looks at Pi_ 12.19.49 Join muesli- [0] (n=muesli_t@141.71.4.178) 12.19.57 # hu? 12.20.11 # you're hereby entrusted with redesigning our gui approach 12.20.16 # :-) 12.20.24 # * preglow hands Pi_ the hat of Much Work 12.21.00 # and i thlought i would get round to painting my bathroom today... 12.21.25 # no, you'll be painting rockbox 12.21.40 # why are people affraid of using their real names in the wiki?? 12.22.02 # and receive the official Flaming Of The Year from rockbox oldtimers 12.22.21 # Bagder: http://130.89.160.166/Gigabeat/from.asm.bz2 12.22.29 # saa[b_r]ider: many people are afraid of using their real names on internet, for some reason 12.22.50 Join Moos [0] (i=DrMoos@m53.net81-66-159.noos.fr) 12.22.56 # Hello here! 12.22.57 # the eternal fear of being raped up the ass by some cyber terrorist that visits people he knows the name of 12.22.58 # I put a lot more comments in there. Maybe we can figure out which devices are connected where from the data that is send to the different ports 12.23.02 # moning Moos 12.23.19 # saa[b_r]ider: maybe they have some unfinished busines with the mafia.. 12.23.49 # I must admit... I didn't want to link between my screen name and my real name, but I didn't care about it too much later... 12.24.29 # but now I'm reading about people who make really cool WPSs who end up not posting at RB because they don't want to use their names, which is a shame really 12.24.46 # at least there's a link to the MR gallaries 12.24.57 # it's a mystery to me 12.25.07 # people are strange 12.25.13 # better stay working with machines 12.25.28 # :-) 12.26.41 Quit NibbIer (Read error: 110 (Connection timed out)) 12.27.46 # and i just plain don't care 12.28.00 # anyone who really wants to know who i am, will find out 12.28.13 # i'm not about to start bouncing from russia to use irc 12.30.15 # Heh. I must say I thought twice. 12.30.21 # Bagder: the assembly code is quite useful, don't you think? 12.30.30 # aliask: why? 12.30.31 # It was the first time I had made any connection between aliask and Will Robertson. 12.30.43 # No idea, I guess I was used to anonymity. 12.30.55 # ok 12.31.08 # (The truth is I'm a super 1337 h4x0r) 12.31.13 # haha 12.31.16 # Zero Cool! 12.31.23 # hahaha :) 12.32.15 # i was very used to using handles from bbses, irc and the scene myself, but for some reason i never thought twice about disclosing my name 12.32.19 # perhaps i'm just plain stupid 12.33.07 # i'll be the first against the wall when the cyber terrorists come 12.34.13 # o noes~1! 12.35.59 # If only VIM could tell me where my semicolon is missing... 12.37.20 Join NibbIer [0] (n=sven@port-212-202-193-50.dynamic.qsc.de) 12.38.16 Join ashridah [0] (i=ashridah@220-253-122-239.VIC.netspace.net.au) 12.42.17 Quit zamzon ("CGI:IRC") 12.46.51 # * amiconn wonders why the status bar needs to be configurable 12.47.11 # Adapting the standard view depending on the display should be enough 12.47.29 # ...and it would avoid code size increase 12.47.44 # for the tastes of anyone 12.47.58 # Like LCD_HEIGHT/8 or something? 12.48.11 # I'm remenbering it was in the Timid todo list 12.48.13 Quit _FireFly_ ("Leaving") 12.53.09 # i don't think it *needs* to be configurable 12.53.40 # no needs but that could be a *plus* 12.54.03 # It's not always a plus to make something configurable 12.54.17 # appearance i think is a plus to make configurable 12.54.40 # as it is, only wps is configurable, and i'd like to see that extended 12.54.43 # that depand, for people desn't wanted change it, the defaut one is here 12.55.15 # perhaps it's possible to retain the good old static code based one for archos units 12.56.33 # i don't think the status bar needs to be configurable, but it sure would be nice to be able to change the appearance of it 12.56.46 # at least the color scheme 12.56.50 # maybe even the icons 12.57.08 # to fit into a future "theme" concept 12.57.35 # yes the evolution due to the new targets 12.58.11 # but i still think that a color lcd is a waste on an mp3 player... 12.58.47 # indeed, music no need color 12.59.17 Join PaulJ [0] (n=PaulJ@vpn-3037.gwdg.de) 12.59.35 # what isn't a waste these days? 12.59.46 # hehe :) 13.00.10 # it's nifty and pretty, and that's enough for people to want it 13.02.41 Quit edx (Read error: 110 (Connection timed out)) 13.02.47 # it's less of a waste than some things i could mention 13.02.52 # at least it opens for new applications 13.03.04 # indeed 13.03.19 # can anyone tel me if the colors pach works? 13.03.39 # the one at geocities? 13.04.13 # yes, i also submited it on source forge 13.04.25 Quit DreamTactix291 (Read error: 110 (Connection timed out)) 13.04.55 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 13.05.06 # cool, cause I can't access geocities, because I'm behind the great wall of china ;) 13.05.14 # For me, coding for iriver is less fun than coding for archos. 13.05.16 # I'll get it from SF now 13.05.29 # Coding for gigabeat would be too boring for me 13.05.44 # The CPU is just overkill 13.05.56 # http://www.hezmatt.org/~mpalmer/blog/general/tabs_are_for_indents_damnit.html 13.06.04 # now there's a good rant 13.06.18 # Pi_: you don't need the #ifdef HAVE_LCD_COLOR for the #include "lcd.h" 13.07.10 # Pi_: the indentation looks weird to me 13.07.19 # amiconn: did you already choose your next target? 13.08.00 # pi: what's the patch's code? 13.08.04 # i just put the #ifdef HAVE_LCD_COLOR round all the code i edited 13.08.30 # 1396750 i think 13.08.35 # amiconn: there are plenty of areas where that cpu would be considered limited 13.08.56 # some of them applicable to a dap 13.09.23 # Pi: can't find it in http://www.rockbox.org/patches.shtml 13.10.42 # it is here http://sourceforge.net/tracker/index.php?func=detail&aid=1394746&group_id=44306&atid=439120 13.11.01 # thanks, Pi: you're in glasgow right? 13.11.02 # saa[b_r]ider: that page only updates twice a day so it lacks the most recent patches 13.11.15 # edinburgh 13.11.17 # Badger: nice to know :) 13.11.46 # I was in edinburgh two summers ago (wow time flies!!!) 13.12.18 # Pi_: why all these bool functions? 13.12.47 # the ones in settings_menu? 13.12.51 # you can use the menu without function pointers and instead check which item that was chosen 13.13.14 # i copied that part of the code from san 13.13.36 Join _FireFly_ [0] (n=FireFly@pd95b7c08.dip0.t-ipconnect.de) 13.14.23 # I think paprica might've worked on that 13.15.06 # i did think it was a bit cumbersum but i don't know enough about it to improve it 13.16.58 # Pi: how's princess street this time of the year :) 13.17.46 # i raerly mke it that far into town, though i expect they whill have cleaned it up after the street party by now 13.18.27 # (I rarely make) 13.19.29 Quit hshah ("Leaving") 13.19.39 # pi: p0 or p1 for your patch? 13.19.46 # ? 13.20.22 # (I realy don't understand how the pach sustem works) 13.20.24 # hmm 13.20.28 # mpeg4 lossless coding 13.20.32 # seems we have a new codec to support :) 13.20.38 # no problem, I'll try and see 13.20.58 # saa[b_r]ider: i used -p0 13.20.58 # preglow: the 5G video ipod supports it, doesn't it? 13.21.11 # cool 13.24.44 # saa[b_r]ider: i have no idea, i got the impression the standard is rather new 13.25.21 # mpeg 4 video? 13.25.24 # audio 13.25.31 # as far as i can see, it's very similar to flac 13.25.47 # preglow: linuxstb was working on it already I believe :) 13.27.13 # 2005-12-31: 11.12.31 # * linuxstb is wasting his time with the new MPEG-4 Lossless Codec 13.27.17 # hahah 13.27.48 # Ouch! The iRiver usb cord just zapped me! 13.28.27 # That's not supposed to happen at 500ma/5v? 13.28.37 # stop playing with that remote usb zapper thing preglow! 13.28.42 Join ST [0] (i=STx@c211-30-160-92.blktn3.nsw.optusnet.com.au) 13.29.06 # aliask: it does happen 13.29.16 # * preglow hides his rockox backdoor access key 13.29.19 # aliask: that's assuming there is infact a voltage clamp at 500ma, and you can feel far less than 500ma anyway 13.29.35 # you can feel 500ma _very_ well 13.29.36 # heh 13.30.00 # Really? Wow. I guess I won't go sticking the USB cord into my arm again... (I was actually doing that...) 13.30.00 # anyway, what you felt was not 500ma, i can guarantee that 13.30.11 # PaulJ: have you tried color.patch? 13.30.12 # you're not that conductive, hehe 13.30.23 # aliask: whatever you do, don't let it penetrate the skin at both ends 13.30.52 # Both ends? Like the small end and the big end? 13.31.10 # no, i mean the cable and ground 13.31.26 # Why, will my computer fry me? 13.31.33 # saa[b_r]ider: yes, it semms to work, and the color is kept after a reboot. 13.31.49 # well, no, but 500ma is more than enough to stop the heart if it crosses the chest 13.31.52 # it's the current that does it 13.32.16 # Even for maximum complexity, the CPU load of the MPEG-4 ALS reference decoder is only around 20-25%, which in return means that file based decoding is at least 4-5 times faster than real-time. 13.32.20 # ..... 13.32.21 # 1.7ghz cpu... 13.32.25 # paulJ: I'm using "patch --binary -p0 < color.patch" but it's not telling me that it patched anything 13.33.00 # preglow: when i read that i thought you were testing it on a iriver... 13.33.13 # they must have coded the decoder in interpreted basic 13.33.22 # Hah. 13.33.24 # saa[b_r]ider: are you in /rockbox ? 13.33.26 # or been gargling whisky while testing 13.33.27 # aliask: shorting it back to itself on your finger wouldn't be that bad tho. 13.33.47 # ah... wait... the patch I downloaded is corrupted 13.33.53 # I can't feel it on the tip of my finger, only the underside of my arm. 13.34.29 # ok, patched :) 13.34.30 # aliask:question... WHY? 13.34.49 # aliask: it'll depend how thin the skin is, sure 13.35.02 # fingers have calluses, which increases the resistance 13.35.03 # Pi_: Beats doing nothing. 13.35.09 # but yeah. "doctor, it hurts when i do this" " then don't do it" 13.35.46 # One of my friends (since dropped out of school) once sucked on mains power (240v) because he was bored. 13.36.18 # ... I need smarter friends. 13.36.23 # saa[b_r]ider> or PaulJ: can u pls upload ur build with the new colour patch for me somewhere, im new, and not into the patching yet 13.36.23 # brilliant 13.36.30 # he's hanged himself for fun since then, i assume? 13.36.32 # i hope it burned his tongue off 13.36.54 # That's not very nice ash... 13.36.57 # http://www.geocities.com/running_amar/rockbox.zip 13.37.10 # aliask: anyone that stupid isn't someone i want to talk to :) 13.37.13 # Pi: which is that 13.37.18 # Fair enough :_ 13.37.19 # ST, sure... let it finish first. but I'm no pro either, you should try it your self as well, or just use pi's link :) 13.37.22 # *:) 13.37.35 # ok thx pi 13.38.08 # PaulJ/other H300 users: have you tried the "add color to WPS" patch? 13.38.45 # i haven't 13.39.04 # I'm waiting for a more substantial patch, like for GIFs or PNGs in the WPS. 13.39.09 # is that the one by Sanitarium on mistic? 13.39.21 # no, 13.39.49 # link pls? 13.39.52 # megavolt (I beleive) from RB... it supposedly adds color to different lines 13.40.27 # I think it requires configuring the WPS file, but there isn't a complete guide for it... (looking for the link) 13.40.55 # Pi: which build is the one u posted? date? 13.41.14 # todays (i think) 13.41.19 # nice 13.41.31 # http://sourceforge.net/tracker/index.php?func=detail&aid=1380389&group_id=44306&atid=439120 13.41.34 # i am not updating the link daily 13.42.09 # its all good, there havent been many changes lately anyway 13.42.40 Quit akaidiot (Read error: 110 (Connection timed out)) 13.45.27 Quit muesli- (Read error: 104 (Connection reset by peer)) 13.48.31 Quit ST () 13.48.53 # Pi: sweet :) 13.50.30 *** Saving seen data "./dancer.seen" 13.53.51 Quit olisto (Remote closed the connection) 13.53.57 # Pi: seems stable. 13.54.19 # glad to hear it 13.58.39 # so can we add more colors into color.patch following this code?: 13.58.39 # +bool background_color_lime(void) 13.58.39 # +{ 13.58.39 DBUG Enqueued KICK saa[b_r]ider 13.58.39 # + global_settings.bg = LCD_RGBPACK(0,255,0); 13.58.39 # + return true; 13.58.39 *** Alert Mode level 1 13.58.39 # + 13.58.41 # +} 13.58.43 # + 13.59.20 # linusN: how would you suggest i change the menus? 13.59.37 Join ST [0] (i=STx@c211-30-160-92.blktn3.nsw.optusnet.com.au) 14.00.29 # have u guys seen http://sourceforge.net/tracker/index.php?func=detail&aid=1374953&group_id=44306&atid=439120 and https://sourceforge.net/tracker/index.php?func=detail&aid=1387447&group_id=44306&atid=439120 >> both with non-lcd remote function? 14.01.00 # saab: you will need to add a comand to the menue below as well 14.02.29 # Pi: alright. makes life easier than having to recompile everytime I decide to change the colors! 14.04.03 # That was just a bit of fun before this was made 14.06.18 # Pi: btw, did you get my pm? 14.06.47 # oh just noticed it 14.08.40 *** Alert Mode OFF 14.08.41 Join muesli- [0] (n=muesli_t@141.71.4.178) 14.25.14 # Pi: that build u posted works fine here 14.26.42 Quit ST () 14.27.45 Quit aliask ("Chatzilla 0.9.69 [Firefox 1.5/2005111116]") 14.29.27 Join Minusandb [0] (n=bob@lns-bzn-50f-81-56-233-117.adsl.proxad.net) 14.39.09 Quit Minusandb (Read error: 104 (Connection reset by peer)) 14.39.43 Join Zak1392 [0] (n=zkeeping@CPE-144-137-207-85.sa.bigpond.net.au) 14.39.47 Nick NibbIer is now known as Nibbler (n=sven@port-212-202-193-50.dynamic.qsc.de) 14.42.43 Join edx [0] (i=edx@p54A85605.dip.t-dialin.net) 14.43.20 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 14.45.51 # hey guys 14.48.45 # thanks for the warm welcome 14.48.58 Join Xerion [0] (n=xerion@cp198589-a.landg1.lb.home.nl) 14.49.02 # hehe 14.49.04 # we don't do welcomes 14.49.27 # consider yourself lucky no one punched you 14.49.40 Quit Xerion (Client Quit) 14.49.51 Join Xerion [0] (i=XERiON@cp198589-a.landg1.lb.home.nl) 14.50.22 # lol 14.51.07 # hola Zak 14.52.27 # hola 14.55.36 Join Mongey [0] (n=mongeyc@213-94-252-135.b-ras1.dbn.dublin.eircom.net) 15.01.15 Join webguest34 [0] (n=864c0325@labb.contactor.se) 15.02.23 Quit Zak1392 () 15.09.29 Quit PaulJ (Read error: 110 (Connection timed out)) 15.10.41 Join Sandking [0] (n=Sandking@ogorek.akron.net.pl) 15.12.54 # i just set polish as a language for my iriver and i see that not all things are translated - i could do it but do i have to have any programming knowledge? 15.13.09 # nope 15.13.26 # too bad 15.13.37 # http://www.rockbox.org/twiki/bin/view/Main/HowtoUpdateLangfile 15.14.25 # not a rockbox runtime database! - i got this again - i think i've deleted it by mistake - there was a really easy way to recreate it because someone told me about it here - any ideas? 15.15.39 # thx for the link - i hope i'll improve the file :] 15.19.23 # that would be nice 15.19.25 # hmm 15.19.32 # perhaps i should sync the norwegian one for a small break 15.21.04 # any ideas preglow about how to create this databasE? 15.21.25 # no idea 15.21.28 # i've never used it 15.22.23 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 15.22.48 Quit _FireFly_ ("Leaving") 15.22.58 # i just like the idea of rating songs and believing that they really are played more often :D 15.23.07 # hehe 15.23.11 # but wtf 15.23.58 # linusN: are you still around? 15.24.06 # whatever encoding i tell vim to use, it still displays corrupted characters 15.24.15 # bah... 15.24.30 Join matsl [0] (n=matsl@1-1-4-2a.mal.sth.bostream.se) 15.26.49 # man, these lang files are getting long 15.29.07 Quit Sandking () 15.35.37 Join Jungti1234 [0] (n=jungti12@58.77.81.144) 15.36.43 # hello? 15.36.49 # hi? 15.36.53 # http://cafe.naver.com/iriverh300/1129 15.38.10 Part webguest34 15.43.14 # funny signs i cant read ;) 15.43.26 # what is this "Screen frozen!" messagE? 15.43.30 # ^^; 15.43.32 # It's an 'improved' viewer 15.43.34 # sounds kind of... strange... 15.45.53 Quit joshn_454 ("KVIrc 3.2.0 'Realia'") 15.47.50 Join ataxy [0] (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 15.48.05 Quit Mongey (Read error: 104 (Connection reset by peer)) 15.50.32 *** Saving seen data "./dancer.seen" 15.50.53 Quit akaidiot (Read error: 110 (Connection timed out)) 15.52.38 Quit muesli- ("ich will Khe!!!") 15.52.42 Quit DJDD_ ("Trillian (http://www.ceruleanstudios.com") 15.53.06 # preglow: "Screen frozen" is for the FM radio screen 15.53.27 # yes, but what function? 15.53.31 # Freezing the screen avoids interference from LCD updates 15.53.36 # right 15.55.15 # which LCD ? 15.55.19 # main or the remote 15.55.26 Join Hans|away [0] (n=maerlyn@p5081A516.dip0.t-ipconnect.de) 15.55.32 # and h100 or h300 15.55.41 # probably both 15.56.03 # but i'd guess the main lcd is the biggest contributor 15.56.37 # but the main lcd on h300 isn't updated most of the time 15.56.53 # i.e. when the backlight is off 15.57.01 # it's probably not a h300 only feature 15.57.45 # w8, are we talking about a patch or about existing feature ? 15.57.52 # existing 15.58.20 # ah, then sorry 15.58.31 # It's for all platforms with FM radio 15.59.10 # The target most susceptible for that kind of interference is the Ondio FM, especially the one with the old Samsung FM chip 15.59.34 # (and of course the H1x0's with the ticking remote problem) 15.59.37 # ah, amiconn, is there a way to change the default osc. for the FM chip on irivers ? 15.59.47 # ?? 16.00.25 # iirc you said that on your iriver h1x0 the fm radio has worse reception than you ondio fm (which is with the same philips tuner) 16.00.41 # yes 16.00.43 # http://www.rockbox.org/twiki/bin/view/Main/PluginRoboracers -.... hu, whats that? is it working(DL-Link)? 16.00.46 # It's less sensitive 16.00.55 # and the reason for this is the worse osc in the irivers 16.01.10 # The oscillator has nothing to do with it 16.01.19 # The osc frequency isn't used directly 16.01.21 # the h1xx FM ist really bad 16.01.31 # I think the antenna coupling is worse 16.01.32 Nick Hans|away is now known as Hansmaulwurf (n=maerlyn@p5081A516.dip0.t-ipconnect.de) 16.01.39 # bad :( 16.03.06 Quit ataxy (Remote closed the connection) 16.03.55 Join ataxy [0] (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 16.05.57 Join UmInAsHoE [0] (n=UmI@81-179-201-99.dsl.pipex.com) 16.23.06 # why oh why must we use a language with no equivalent word for 'default' 16.23.52 # heh, preglow :) 16.23.57 # standardvärde ? 16.25.22 # i 16.25.29 # i'm trying to translate 'default codepage' 16.25.38 # i could use 'standard', i guess 16.26.27 # what kinda of knowledge is required to help the RockBox project (namely programming) ? 16.26.36 # you need to know c programming 16.26.43 # i know C 16.27.02 # then hooray! 16.27.02 # Then welcome! 16.27.10 # :) 16.27.13 # lol 16.27.13 Quit Moos (Read error: 104 (Connection reset by peer)) 16.27.14 Join DrMoos [0] (i=DrMoos@m53.net81-66-159.noos.fr) 16.28.27 # well i seen one patch was written in ASM, what type of ASM is that cause it cant be x86 16.28.54 # UmInAsHoE it depends on the player 16.28.56 # UmInAsHoE: we have Coldfire, SH1 and ARM cpus 16.28.58 # iriver's is m68k 16.29.12 # i got a h3xx so i would be working on that 16.29.29 # UmInAsHoE: what would you like to work on? 16.30.09 # erm not sure, browsing through the Feature requests atm 16.31.10 # what are you all working on at the moment? 16.36.59 # ipod port and eq 16.37.36 # cool ^_^ 16.37.57 Join _FireFly_ [0] (n=FireFly@p54A468FE.dip.t-dialin.net) 16.37.57 Join FireFly_ [0] (n=FireFly@p54A468FE.dip.t-dialin.net) 16.38.58 # if any of them ever finish, yes 16.39.13 # heh lol 16.40.33 # I'm working on a Toshiba Gigabeat port, but it's in a very early stage 16.41.12 # UmInAsHoE: If you can't think of anything see my wish list :) http://www.rockbox.org/twiki/bin/view/Main/MarcoenHirschberg 16.42.11 # one sec reading through some of the RockBox source 16.44.36 Join webguest95 [0] (n=3a4ffa23@labb.contactor.se) 16.44.45 Quit webguest95 (Client Quit) 16.45.53 Join Dodozi [0] (n=3a4ffa23@labb.contactor.se) 16.46.15 # k 16.47.21 Join Matze41 [0] (i=Miranda@p5484F99E.dip.t-dialin.net) 16.47.26 # 안녕하세요!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! 16.47.30 # YES! 16.47.31 # -_-; 16.47.31 # AGREED! 16.47.56 # ?? 16.48.03 # how about a JPEG/BMP/PNG/GIF viewer :) 16.48.20 # just my standard enthusiastic response to gibberish followed by a million exclamation marks 16.48.21 # saa[b_r]ider: then extend the jpeg-viewer ;) 16.48.59 # 会えて嬉しいです。 16.49.03 # amen! 16.49.10 # Dodozi: english please 16.49.11 # amen! 16.49.14 # dodozi 16.49.18 # what are u saying? 16.49.29 # I would if I knew how to... I try to help in every way I can :) only UmInAsHoE was looking for something to do, so I thought I'd suggest 16.49.30 Quit Dodozi (Client Quit) 16.50.01 # and to whom? 16.50.01 # oh well... we scared him off 16.50.21 # saabrider: grayscale-jpeg for Non-Colored models (cause Grayscalelib doesnt work on H300 series), BMP and an alpha version of gif is already there 16.50.42 # o.O 16.51.32 # Because I want something like that, too, i think i probably will see what I'm able to code about Color-JPEG. PNG should be quite easy I think 16.51.32 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 16.52.11 # there's also the new "viewbox" project which hasn't been added to CVS... which has a buggy GIF viewer and is supposed to support jpeg as well afaik 16.53.05 # jpeg? 16.53.08 # well let me read... 16.53.45 Join gtkspert [0] (n=gtkspert@203-59-207-7.dyn.iinet.net.au) 16.54.16 # do you have a link to jpeg decoder? 16.54.34 # libjpeg ?? 16.54.44 # oh no i mean for rockbox :) 16.55.29 Quit gtkspert (Client Quit) 16.57.38 Quit UmInAsHoE ("Yes I Quit. No You May Not Follow Me.") 16.57.43 # Matze41: have a look at this... http://www.misticriver.net/showthread.php?t=34392 16.58.02 # I've already seen this 16.58.05 # there is no jpeg at all 16.58.47 # even the "beta" at the end? 16.59.38 # well let me have a look... 17.00.21 # add these lines to the "viewers.config" 17.00.21 # bmp,viewers/viewbox,00 FF 89 89 76 00 17.00.21 # gif,viewers/viewbox,00 FF 89 89 76 00 17.00.21 DBUG Enqueued KICK Matze41 17.00.21 # and copy "viewbox.rock" to "viewers" folder... 17.00.26 # this is last post by eli sherer 17.00.32 # so no jpeg, sry 17.01.17 # yeah true... but he was on his way to add it... 17.01.25 # he never said this 17.01.27 # did he? 17.01.31 # hang on... 17.01.37 # Grayscale JPEG on H300 should be almost trivial 17.01.59 # souleata: I think instead of a million different filetype viewers, there should be one image viewer. 17.01.59 # (re-)adding colour support to the JPEG decoder is a bit more work 17.02.10 # eli: that is what i'm aiming for...wait for the next version of "viewbox" 17.02.21 # amiconn: yes it should be... I tried porting grayscalelib in 5 minutes... but it is very undocumented and i think a little bit complex so i had no success 17.02.26 # ah 17.02.54 # It wouldn't make sense to port the grayscale lib to h300 17.03.06 # the BMP viewer was his first, then this GIF port.... 17.03.09 # well, saabrider: i think this is just for "one imageviewer for all files" 17.03.13 # yep 17.03.15 # The H300 LCD is able to display enough shades natively 17.03.16 # then he disappeared 17.03.23 # Gifreader already includes bmp viewer 17.03.42 # ok.. 17.03.43 # http://blogfiles16.naver.net/data12/2005/11/2/143/5000_1-ss_sakura.jpg 17.04.05 # amiconn: yes i didnt really try to port grayscale lib... and if so, it would just be an alias with 32 grayscales to the "normal" lcd-calls 17.04.18 # jungti: can you buy a can of coke with that :) 17.04.32 # but grayscale lib has this initializing and many points i didnt really understand in 5 minutes :) so maybe another time 17.04.36 # saab: heh 17.04.53 # http://www.rockbox.org/twiki/bin/view/Main/GraphicsAPI 17.06.06 # What is necessary is to implement lcd_gray_bitmap[_part]() and lcd_color_bitmap[_part] in the plugin library. Should be very simple compared to what the grayscale lib does 17.07.15 Ctcp Ignored 2 channel CTCP requests in 3 hours and 1 minute at the last flood 17.07.15 # * amiconn wrote the grayscale library :) 17.07.36 # ah :) 17.07.36 # k 17.07.40 Quit Kohlrabi (Read error: 104 (Connection reset by peer)) 17.07.48 # but why do you think we need this functions in plugin library? 17.07.59 # we could directly work with framebuffer... 17.08.00 # Because we don't need them in the core 17.08.15 # Yes, but this way it is easier to re-use stuff 17.08.23 # good night 17.08.26 # yeah thats right 17.08.35 Quit Jungti1234 () 17.08.41 # jungti: 18 cans ... bah, he left 17.08.53 Join Kohlrabi [0] (n=Kohlrabi@dslb-082-083-129-122.pools.arcor-ip.net) 17.09.10 # saa[b_r]ider: There is one reason against an all-in-one image viewer, and that reason is resources 17.09.36 # Not *that* important on the newer targets, but very important on archos 17.09.54 # Having one viewer per format also reduces complexity 17.09.57 # hm i think we should make different imageviewers for the targetclasses 17.10.06 # so one for grayscale 17.10.10 # one for b&w 17.10.12 # and one for colour 17.10.14 # amiconn: I take your word for it. and I respect the elderly :) 17.10.37 # (aka archos) 17.11.00 # because it is really no waste on resources to call different decoding&viewing functions for every filetype... 17.11.08 # you could define an image-viewer-framework 17.11.20 # I don't really mind having seperate viewers.. I don't want to make it harder for you guys to code, by any means 17.11.29 # Matze41: Still all different functions would be part of the binary 17.11.47 # yes 17.12.03 # Check the size of jpeg.rock for archos and compare with the plugin ram size... 17.12.08 # is the whole binary hold in memory when executed? 17.12.42 # and the otherside you can*t have 2 or more image-files open at the same time afaik in the viewer 17.12.43 # Perhaps it would be possible to use some kind of framework 17.12.45 # : 17.13.06 # or maybe the question has to be: why is the binary so fucking big? The corefunction are only CALLED i think? or are they "copied" into the binary? 17.13.15 # which defines the "ui" and each new viewer has only to reimplement the image-loading 17.13.37 # Have a number of image decoder functions, and a generic viewer framework. Link this either into separate viewers or one all-in-one viewer 17.14.05 # hm 17.14.25 # Matze41: The biggest part _is_ the jpeg decoder itself 17.14.43 # hm 17.14.47 # ...even though it only decodes luma, and doesn't support all jpeg variants 17.14.57 # (e.g. no progressive support) 17.14.58 # the sourcecode is few times smaller than the binary i think? 17.15.15 # and normally sourcecode is with comments etc.. 17.15.40 # because on the targest there is no malloc/free so each buffer is a static buffer 17.15.52 # and these static-buffer makes the binary big afaik 17.15.52 # is it really only the decoderthing itself, not a "library" or something like that that get linked into the binary? 17.16.09 # well that makes sence 17.16.09 # jpeg.c: 66816 bytes. Archos recorder jpeg.rock: 26084 bytes 17.16.20 # oh 17.16.22 # k 17.16.34 # The archos plugin ram is 32KB (32768 bytes) 17.16.42 # why only this low? 17.16.48 # i think it has 8mb ram? 17.16.48 Part Pi_ 17.16.57 # It has 2MB RAM 17.16.59 # oh 17.17.01 # it needs all of those buffers? None can be shared? 17.17.04 # well okay :) 17.17.22 # It would be possible to increase the plugin RAM area 17.17.27 # and rockbox itself nearly needs 2mb? 17.17.29 Join webguest24 [0] (n=414a01bd@labb.contactor.se) 17.17.44 # yes i think this would be neccessary for some other "nice" plugins too 17.17.57 # 32kbyte is really nothing 17.18.04 # afaik onw of the biggest parts is the audio buffer 17.18.11 # The problem is that this RAM is set aside *permanently* and can't be used for buffering music 17.18.21 # amiconn, how's the wake up feature coming on ? 17.18.24 # firefly: recorder has hardware decoder 17.18.32 # so no audio buffer in ram i think 17.18.43 # then no crossfade is possible affaik 17.18.45 # -> battery runtime is decreased, as it needs to rebuffer more often 17.19.01 # Matze41: Of course it has to buffer audio data in RAM 17.19.16 # okay maybe audio data is wrong word 17.19.25 # the decoder is faster then reading from disk 17.19.30 # oehm, it buffers compressed data 17.20.00 # audiodata for me is uncompressed pcm 17.20.14 # FireFly_: Of course not. Otherwise it wouldn't even be possible to rebuffer fast enough 17.20.26 # ok 17.20.33 # Matze41: I said audio data, not PCM audio data 17.20.40 # yeah okay sry i misunderstood 17.21.21 # Of course rockbox buffers compressed data (exclusively on hwcodec platforms, i.e. archos, and mostly on swcodec platforms, i.e. iriver and others) 17.21.42 # yeah okay 17.21.59 # It's more efficient this way. 17.21.59 # hmm btw... archos has a harddrive, too? 17.22.13 # Didn't see if someone already considered the possibility of the viewer plugin being able to dynamically load image codecs into part of the plugin ram? would add build time complexity, but not infeasible? 17.22.23 # with only 2mb buffer it has to spin up VERY more often than iriver one's i think? 17.22.33 # lostlogic: plugins for plugins??? 17.22.49 # Matze41: Yes. 17.23.05 # * webguest24 raises hand in the air, high 17.23.18 # That's why we want to keep RAM usage for other purposes than the main buffer as low as possible 17.23.19 # essentially, yes -- but much simplified, because there need not be callbacks, just a couple of callins 17.23.33 # amiconn: okay that makes sence 17.23.45 # and plugins should be callable with music playing 17.24.05 # Yes, and generally they are 17.24.09 # yes 17.24.16 # okay 17.24.20 # Some have to stop playback, because they take over the main buffer 17.24.27 # (like the jpeg viewer) 17.24.31 # oh it does? 17.24.50 # well okay makes sence 17.24.59 # 200kbyte jpeg files or so... cant be decoded in 32kbyte 17.25.22 Part webguest24 17.25.24 # Yes. Very much impossible to fit the grayscale image buffer (28KB) plus the jpeg file plus the decoded image(s) into 6KB ;) 17.25.33 # lol yeah 17.26.07 # okay.. 17.26.17 # On iRiver it should be possible to do jpeg viewer in plugin ram, on small images at least...? 17.26.22 Join Lear [0] (n=chatzill@h247n1c1o285.bredband.skanova.com) 17.26.38 # but i think mainproblem was not the problem about an all-type imageviewer 17.26.59 # because we're talking about color jpeg/gif/bmp/png 17.27.21 # and color-targets ( i think its only the iriver?) have enough memory for that 17.27.35 # ipod color 17.27.38 # Hmm, also, if the jpeg viewer takes over the mainbuffer any way, why not dynamcally allocate it's other buffers there as well to reduce the .rock size 17.27.39 # and ipod video 17.27.40 # lastlogic: yes irivers pluginram is big enough for anything i think 17.27.40 # Iriver H300 plus iPod color and Nao atm 17.27.45 # *Nano 17.27.57 # oh 17.28.15 # Ah, yes, and iPod video 17.29.10 # how about those, do they have "enough" ram? (well, the quotes because no player has enough ram... buffering the whole harddrive content would be enough^^) 17.29.52 # because i think we should make the imagecolorviewer in one file 17.30.17 # maybe different decoders can even use same variables? they dont run same time... 17.30.53 Quit FireFly_ ("Leaving") 17.30.58 Quit _FireFly_ ("Leaving") 17.31.04 Join _FireFly_ [0] (n=FireFly@p54A468FE.dip.t-dialin.net) 17.33.07 Quit Zagor ("Client exiting") 17.34.10 # ipod video even has 64MB of ram 17.35.15 # * preglow stares at the latest ml post 17.38.59 # Do you mean the one by Manuel Dejonghe? 17.40.17 # <_FireFly_> lol 17.40.31 # <_FireFly_> this mail was send to the wrong address 17.40.40 # happy new day 17.41.11 # markun: 64 ? 17.41.20 # markun: oh for buferring that's nice 17.43.26 # :) 17.43.31 # aeh i have one question 17.43.38 # where is the individual memorysize defined? 17.43.50 # the config-model.h only contain codec and plugin buffer size, not ram size 17.44.47 # <_FireFly_> Matze41: for archos the size is asked/given when tools/configure is executed 17.45.42 # does the sim have lcd color now ? 17.46.00 # okay so configure does this... k i'll have a look at it later. now i'm going to have a shower :) -> afk 17.46.17 Join Pi_ [0] (n=Pi@80-44-103-155.dynamic.dsl.as9105.com) 17.46.18 # take a shower i mean 17.46.46 # <_FireFly_> mirak: afaik only the win32-sim 17.46.51 Join stevenm [0] (n=stevenm@209.101.154.70) 17.47.18 Nick DrMoos is now known as Moos (i=DrMoos@m53.net81-66-159.noos.fr) 17.47.40 # _FireFly_: hum, is that easy to have a build environnement for windows on linux ? 17.47.50 Quit stevenm (Read error: 104 (Connection reset by peer)) 17.48.43 Part Pi_ 17.48.59 Join Pi_ [0] (n=Pi@80-44-103-155.dynamic.dsl.as9105.com) 17.49.07 Join stevenm [0] (n=stevenm@209.101.154.70) 17.49.38 Nick ataxy is now known as ataxyab (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 17.50.10 Quit stevenm (Read error: 104 (Connection reset by peer)) 17.50.33 Quit Pi_ ("Trillian (http://www.ceruleanstudios.com") 17.50.36 *** Saving seen data "./dancer.seen" 17.51.07 Join Pi [0] (n=Pi@80-44-103-155.dynamic.dsl.as9105.com) 17.51.22 # amiconn: yeah, that 17.51.57 # i don't often see people plan casino nights per mailing list 17.52.56 # mirak: If you use debian, then yes. There's a mingw cross-gcc package available, and the sim works under wine 17.53.13 Part LinusN 17.58.15 Join stevenm [0] (n=stevenm@209.101.154.70) 17.58.52 Quit stevenm (Read error: 104 (Connection reset by peer)) 18.00.02 Join stevenm [0] (n=stevenm@209.101.154.70) 18.04.22 Quit stevenm (Read error: 104 (Connection reset by peer)) 18.04.57 Join ender` [0] (i=ychat@84.52.165.220) 18.05.24 Nick ataxyab is now known as ataxy (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 18.06.46 Join NicoFR [0] (n=nico404@rob92-6-82-231-243-63.fbx.proxad.net) 18.07.39 # amiconn: how to specify the compiler ? 18.07.49 Join LinusN [0] (n=linus@labb.contactor.se) 18.09.39 Join Philip_0729 [0] (n=Philip_0@user-805.l6.c4.dsl.pol.co.uk) 18.10.37 # mirak: It's done automatically if you select to build a win32 sim 18.12.56 # The win32 sim builds on the rockbox.org build server are done the same way afaik 18.13.10 # I have a problem in time.h 18.13.18 # I don't have updated cvs for a long time tough 18.18.29 # linus: are you about? 18.31.28 # /home/karim/Prog/src/rockbox/rockbox/firmware/export/time.h:22:22: features.h: No such file or directory 18.31.36 # /home/karim/Prog/src/rockbox/rockbox/firmware/export/time.h:22:22: features.h: No such file or directory 18.31.56 # something is missing 18.34.48 # amiconn: do you know how to fix that ? 18.35.45 Quit ender1 (Read error: 110 (Connection timed out)) 18.38.40 Quit DreamTactix291 (Read error: 104 (Connection reset by peer)) 18.38.53 Join DreamTactix291 [0] (n=DreamTac@adsl-149-149-95.bna.bellsouth.net) 18.41.09 # oh my god ! there is the clock on H300 :D 18.42.23 # lol 18.43.20 # lostlogic: btw, do you need a map file from me to fix the parser? 18.43.23 # I made a patch so you can put it in WPS too, not sure if the patch is 'right' to get included, or if it will need changes (it's on patchtracker somewhere) 18.44.48 # Lear: yeah, would be most helpful, along with objdump -t output just to double check. lostlogic at lostlogicx.com 18.45.40 # objdump on the elf file? 18.45.48 # on libTremor.a 18.46.12 # hem I want to use timer things 18.46.26 # long* current_tick; that's the api variable 18.46.32 Quit saa[b_r]ider (Read error: 110 (Connection timed out)) 18.46.36 # I have seen there is macros in time.h 18.46.43 # I am not sure how to use that 18.46.48 # separate things 18.46.54 # what kind of timer are you looking for\ 18.48.02 # well I am still on xvid stuff. I want to toy with iram. Actually I have kind of make dummy functions of timing fuctions 18.48.09 # the content is commented 18.48.21 # but everything is there in the code 18.48.39 # timing functions? to count how many ticks the operations are taking? 18.48.49 # yes 18.49.05 # there is a msecond function for exemple that return a double 18.49.08 # struct timeval tv; 18.49.08 # gettimeofday(&tv, 0); 18.49.08 # return((double)tv.tv_sec*1.0e3 + (double)tv.tv_usec*1.0e-3); 18.49.21 # gettimeofday will not work probably 18.49.21 # You could also use the profiling patch I wrote for that... the patch exports it to the plugin API and all... 18.49.43 # what does it do ? 18.49.52 # it ... profiles... 18.50.04 # what is it ? 18.50.09 # question: i use the directory-buffer, but it seems rockbox doenst save it, when i go into the directory-browser after some while, i can move dir-levels up, but not down, do i something wrong? 18.50.16 # lookup software profiling. 18.50.19 Join ender1 [0] (i=ychat@84.52.165.220) 18.50.25 # lostlogic: mail sent. 18.50.27 # ??? 18.51.20 # mirak: you won't get fine grained timing using the get_time related functions and even the current_tick isn't msec precision. the profiling code uses a system time for finer grain and stuff. 18.52.06 # the system time is related to cycles ? 18.52.08 # Lear: Ok, I won't really be able to work on it until about 0:00GMT tonight 18.52.35 # No panic, I have that python script for now... 18.52.50 # mirak: the tick is kept constant length as the clock speed changes. 18.53.00 # lostlogic: do you have link to the patch ? 18.53.15 # it's on patch tracker, look for profiling 18.54.18 Quit ashridah (Read error: 110 (Connection timed out)) 18.56.14 # I have come to the realisation that I am lazy 18.56.41 Join ashridah [0] (i=ashridah@220-253-120-186.VIC.netspace.net.au) 18.58.51 Quit YouCeyE ("Leaving") 18.59.48 # lostlogic: I think I could just use ticks and decode 5 seconds of stream 18.59.50 # lostlogic: you've started work on perlifying the script? 19.00.21 # mirak: that wont help you find out _where_ to optimise 19.00.56 # preglow: I don't see what it's doing 19.01.08 # I don't see how I can use it, I never used that 19.01.46 # maybe you can tell me a bit more 19.03.31 # i haven't had time to try it myself 19.10.27 Quit ender` (Connection timed out) 19.22.45 Join nebratu [0] (n=non@70.134.128.97) 19.25.41 Nick Paprica[bbl] is now known as Paprica (n=Paprica@CBL217-132-98-185.bb.netvision.net.il) 19.25.43 # hi 19.35.33 Quit akaidiot (Connection timed out) 19.41.17 Join actionshrimp [0] (n=NNSCRIPT@host86-136-16-248.range86-136.btcentralplus.com) 19.41.58 Quit nebratu () 19.43.04 # i ham having trouble with the set_option menu, I keep getting "phrase error before int" 19.43.17 # *I AM 19.44.20 Quit ender1 (Read error: 104 (Connection reset by peer)) 19.46.00 Join ender` [0] (i=ychat@84.52.165.220) 19.46.19 # I am using rb->current_ticks to get the time but it seems static 19.46.28 # when the plugin run 19.47.36 # I do : int start_tick= rb->current_tick; 19.47.55 # then : int total_ticks= rb->current_tick - start_tick; 19.49.08 # Add * 19.49.24 # lol ok 19.49.28 # (As in "*rb->current_tick") 19.50.03 # roger 19.50.39 *** Saving seen data "./dancer.seen" 19.52.26 # can anyone see what is wrong with "ret=set_option("Color", *color, int, names, 3, NULL);" 19.55.20 # Looks like it should be "&" rather than "*"... "int" -> "INT". Hard to say more without more code. :) 19.55.59 # thanks ill try that 19.57.03 # great, it worked! 19.57.25 Join muesli__ [0] (i=muesli_t@88.134.37.38) 19.59.18 # re 20.00.40 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 20.02.35 Quit Nibbler (Read error: 110 (Connection timed out)) 20.02.41 Join Nibbler [0] (n=sven@port-212-202-193-50.dynamic.qsc.de) 20.03.20 # ICONST_ATTR IDATA_ATTR hum what practical difference is there between this two iram spaces ? 20.03.33 # Lear: are you there ? 20.04.05 # Both do essentially the same thing 20.04.09 # ICONST_ATTR is for const data 20.04.19 # IDATA_ATTR for non-const initialised data 20.04.35 # IBSS_ATTR for ordinary data 20.04.44 # The reason is that you can't put 'const' and non-const data into the same section, because then gcc would report a section type conflict 20.05.43 # ok 20.05.52 # Huh? Doesn't quite seem to match my experience... 20.06.25 # what's the size of each section ? 20.06.51 # But maybe that does explain it... 20.07.24 # Each *_ATTR you mean? Not limited as such, all three must fit within the reserved area though. 20.08.01 # 4 20.10.55 # I have put the idct into ICODE_ATTR , but it crashes 20.11.40 # void ICODE_ATTR 20.11.40 # idct_int32(short *const block) 20.11.52 # it doesn't work like that 20.12.05 # void myfunction(int hehe) ICODE_ATTR; 20.12.17 # void myfunction(int hehe) { printf("lol %i\n", hehe); } 20.12.31 # that's how you need to do it 20.12.41 # if myfunction(int hehe) ICODE_ATTR; is in the header is it ok ? 20.12.57 # or it must be also in the .c file preglow ? 20.12.57 # And, if you do this in a plugin/codec, you need to copy the section yourself 20.13.12 # ?? 20.13.21 # just have a look at the codecs to see how that part works 20.13.32 # as to the prototype, i actually don't know, i've always had them in the .c file 20.13.43 # try it out and see, i have a nagging suspicion that it needs to be in the .c 20.14.56 # brb 20.16.19 # Having the prototype with *_ATTR in an included .h file is sufficient 20.18.23 Quit Kohlrabi (Read error: 104 (Connection reset by peer)) 20.19.14 # amiconn: I don't understand what the memset is doing 20.19.22 Join Sacro [0] (i=Sacro@adsl-213-249-237-99.karoo.KCOM.COM) 20.19.34 Join Kohlrabi [0] (n=Kohlrabi@dslb-082-083-129-122.pools.arcor-ip.net) 20.19.35 # extern char iramcopy[]; 20.19.43 # what's the extern for ? 20.20.58 Quit ashridah (Read error: 110 (Connection timed out)) 20.21.19 # well 20.21.23 # it's an external variable 20.21.30 # that is, not defined in the current .c file 20.21.33 # I guess it's for having it dynamic or something ? 20.21.40 # I mean the whole memset stuff 20.21.59 # why there is nothing to do for the ICONST N 20.22.00 # ? 20.22.08 # ICONST_ATTR 20.22.22 # it's because of that, yes 20.22.36 # the plugins are just simple binary images, there's no way to tell which part of it contains the const iram data 20.22.52 # so the plugin itself has to copy it, since it's the only thing that knows which parts are meant to be in iram 20.22.54 # The memcpy() has to be done as soon as you use one of (IDATA_ATTR, ICONST_ATTR, ICODE_ATTR) 20.23.11 # The memset() _should_ be done as soon as you use IBSS_ATTR 20.23.56 Join saa[b_r]ider [0] (n=saab_rid@221.223.96.206) 20.24.26 # btw, i had a quick look at making the plugin format a bit more versatile 20.24.44 # and it does indeed seem we have to use something pre-existing unless we want to code our own tools to do the object conversion 20.24.54 # the simplest i can think of is probably a.out or coff 20.25.13 Join ender1 [0] (i=ychat@84.52.165.220) 20.27.03 # amiconn: ok I understand now why it was displaying nothing 20.27.54 # preglow: Why? Do you want plugins to be relocatable? 20.28.16 # for more flexibility 20.28.17 # Imho this would be overkill... 20.28.39 # for example to load more at once without linking a zillion different ones 20.29.18 # we already need to jump around this issue with codecs 20.29.27 Join akaidiota [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 20.31.24 Join mozetti [0] (n=moe@p54A86057.dip.t-dialin.net) 20.33.08 # i agree it's not the number one thing rockbox needs at the moment 20.33.17 # but it would be cool, and i'm willing to bet, sooner or later we will need it 20.33.32 # it takes 821 ticks to decode 20 frames. by putting an array used in idct I gain 12 ticks and got to 809. By putting the idct function into iram I go down from 809 to 790. I gain 19 ticks. 20.33.43 # further splitting codecs into different parts (metadata, loaders) certainly will 20.34.24 # that's really not much 20.35.31 # well, the trick is finding the important stuff 20.35.50 # and profiling is useful there... 20.36.02 # exactly 20.36.13 # probably, but seems complicated for me for now 20.36.31 # I will just note this gains in the code for now 20.36.38 # it isn't very complicated 20.36.55 # I would be interested in knowing wich method is called the most 20.37.02 # I guess profiling can do that ? 20.38.43 # yes 20.38.45 # that's what it does 20.39.25 # but you must put a call to the profilethread method in each function ? 20.39.30 # no 20.39.31 # it's automatic 20.39.40 # that's aspect programming ? 20.39.44 # not really 20.39.54 # I have looked at the diff file 20.39.54 # go have a look at the patch 20.39.58 # I did that 20.40.09 # I did not patched tough 20.43.43 Quit akaidiot (Read error: 110 (Connection timed out)) 20.43.46 # iram is to small :-/ 20.45.03 # sure is 20.45.08 # but still more than enough for some decent speedups 20.49.59 # does the ipod have the same sort of iram area like the iriver 20.50.07 # yes 20.50.21 # it's even 96kb on some models, 128kb on nano and 6g 20.50.22 # 5g 20.50.39 # it's an arm processor, not m68k though? who makes the chip? 20.50.53 # arm is licensed to whoever wants it 20.51.05 # this particular chip is made by portalplayer 20.51.14 # interesting. 20.51.22 # but everyone and his mother has an arm core 20.51.27 # arm based core, that is 20.51.43 # the core itself is very low on transistor count, so it's popular in soc amplications 20.51.45 Quit akaidiota (Read error: 110 (Connection timed out)) 20.51.57 # lol I read it as: everyone and his mother has an arm :) 20.55.18 Quit ender` (Connection timed out) 20.59.01 # can somone translate what "warning: assinment makes integer from pointer without a cast" means 21.00.16 # the line that causes the warning is " global_settings.bg = color_select;" 21.01.45 # color_select is a pointer, bg isn't 21.02.36 # <_FireFly_> *color_select 21.03.03 # so i have to work out how to define color_select as not a pointer 21.04.06 # but what is color_select? how is it used? 21.04.22 # <_FireFly_> or use *color_select to get the value to which the poiner points 21.05.11 # <_FireFly_> *pointer 21.05.59 # it is a menu that picks what the background color shoul be and it should return an unsigned 21.06.45 Quit ataxy ("Chatzilla 0.9.69.1 [Firefox 1.5/2005111116]") 21.07.18 # _FireFly_: didn't work 21.07.19 Join petur [0] (i=petur@d54C1B62E.access.telenet.be) 21.08.11 # <_FireFly_> which type is color_select 21.08.21 # how it is deffined atm "static unsigned color_select(void) {" 21.08.37 # <_FireFly_> ouch 21.08.41 # <_FireFly_> this is an fn 21.08.55 # yes 21.09.01 # <_FireFly_> then call the fn 21.09.02 # <_FireFly_> #;) 21.09.16 # <_FireFly_> and save the return value in the var global_settings.bg ;) 21.09.35 # <_FireFly_> Pi i hope you know how to use functions 21.09.50 # i thought i did 21.10.14 # <_FireFly_> try: global_settings.bg = color_select(); ;) 21.10.16 # and i though that is what i had done 21.10.20 # <_FireFly_> no 21.10.26 # <_FireFly_> you didn't 21.10.33 # wil try again 21.10.50 Quit Maxime (Read error: 104 (Connection reset by peer)) 21.11.31 Join Maxime [0] (n=flemmard@fbx.flemmard.net) 21.11.49 # thanks, that is that compile warning gine, now lets see if the runtime mess up is fixed 21.12.01 # *gone 21.12.21 # <_FireFly_> Pi i think you should lern a bit more about C and how to use functions :) 21.12.50 # <_FireFly_> s/lern/learn 21.13.13 # yes i am learning as i go, (have used pascal and fortran in the past) 21.13.40 # <_FireFly_> afaik is in pascal the same how to use functions 21.13.48 # <_FireFly_> as in C 21.14.33 # you only need brackets after the function name if you are passing a variable 21.14.49 # <_FireFly_> in c you need always () 21.15.48 # so color_select; would have been fine (exept that functions are functions not variables as well) 21.18.36 # that has fixed it thanks 21.20.04 # anyone know with what I can encode in mpeg2 on linux ? 21.20.52 # <_FireFly_> mirak: video or audio 21.21.58 # video 21.22.25 # I need to have a mpeg2 with the good size 21.22.44 # <_FireFly_> you could use memcoder or if you want a gui the avidemux2 21.22.50 # noticed a bug on the H3xx regarding FM radio presets (I think). After entering the name and pressing 'play' to save, i get some strange text (I00 at: 000001) and the only key it responds to is 'play' which causes it to power down immediately (crash?) 21.25.32 # how risky is it to complile my own bootloader?? 21.25.34 # mirak: ffmpeg 21.26.07 # Philip_0729: why'd you want to do that? 21.26.25 # to change the text colour in it 21.26.30 # hahaha 21.26.33 # go ahead 21.26.35 # :D 21.26.43 # if you fuck up, it's your problem 21.27.11 # i realize that... which is why i'm not going to do it... atm 21.27.15 # and the possibility of it happening is most certainly there 21.27.57 Quit Lear ("Chatzilla 0.9.69 [Firefox 1.5/0000000000]") 21.28.17 # Philip_0729: what do you want to change in the bootloader? 21.28.32 # Ah, forget it, i'm blind :) 21.29.20 # it boots so quikly anyway that i don't realy notice it ;) 21.30.30 # oh u can build in a boot wait! so that anyone can see ur text 20 seconds or so :) 21.31.11 # or u can just be happy about rockbox beeing times faster than original firmware 21.32.09 Quit Xerion ("l8r...") 21.33.18 Quit ender1 (Read error: 104 (Connection reset by peer)) 21.33.26 Join ender` [0] (i=ychat@84.52.165.220) 21.33.51 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 21.34.34 # @Matze41: I didn't meant that... i was refering to why i wouldn't risk changing the boot for soething that is only shown for a few seconds 21.34.44 # *something 21.38.19 # Can somebody confirm the FM preset issue (^^) or explain where the 'I00 at: 000001' comes from? Just to prevent that I go bug hunting in the wrong direction... 21.40.21 Join oxygen77 [0] (n=Chris@vau75-6-82-237-174-211.fbx.proxad.net) 21.41.44 # hello 21.43.58 # I have a question on rockboy, is there anyone that could help me on this ? 21.45.01 # just ask 21.45.05 # yup 21.45.10 # thx 21.45.33 # anybody know the sound format used by the emu ? 21.46.13 # I know it is pcm but with which attribute ? 21.46.28 Quit mirak (Remote closed the connection) 21.47.10 # i dont hear sound on it 21.47.40 # yes, that I know too the code for sound is not complete 21.47.59 # my question was in fact more general and is on gnuboy 21.48.40 # 11khz so far, i think 21.48.49 # 16 bit, stereo, probably 21.49.50 # hum this is specified for rockboy in rbsound.c, but I'm wondering if it can be something else 21.50.41 *** Saving seen data "./dancer.seen" 21.53.09 Part Philip_0729 21.54.29 Join JdGordon [0] (n=546d6b5a@labb.contactor.se) 21.54.38 # hey ALL 21.54.48 # damn capslock 21.55.27 Quit JdGordon (Client Quit) 21.57.10 # just wanna give a big thanks to whomever put together the Dev Kit & instructions. Was able to patch & compile my own daily yesterday. Great work 22.00.38 # that would be bluechip 22.00.45 # who is rarely here 22.03.40 Join ashridah [0] (i=ashridah@220-253-122-119.VIC.netspace.net.au) 22.07.12 Part LinusN 22.13.43 Join solexx [0] (n=jrschulz@c214079.adsl.hansenet.de) 22.18.35 Quit akaidiot (Read error: 110 (Connection timed out)) 22.20.56 Join JoeBorn [0] (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) 22.25.59 Quit solexx_ (Connection timed out) 22.32.21 Quit actionshrimp ("a bird in the bush is worth two in your house") 22.32.55 Join actionshrimp [0] (n=NNSCRIPT@host86-136-16-248.range86-136.btcentralplus.com) 22.36.33 # well i'm just reading through the viewbox from eli sherer, especially the gif code: why does he use sizeof(unsigned short) as an argument for the length of a read call? Just writing 2 would be better because it is a constant on EVERY system. This has to be 2 even on systems where an unsigned short would be 4bytes or so :) 22.37.01 # i mean the readings out of the gif-header for example 22.37.40 Part oxygen77 ("Leaving") 22.40.25 # Writing sizeof(unsigned short) is better because it shows where the value comes from 22.40.34 # A plain 2 can be just about anything 22.41.01 # The generated code is exactly the same 22.45.53 # does anyone know how to use the menu without function pointers and check which item was chosen 22.57.05 Quit ashridah ("Leaving") 22.57.30 # what do you do 22.57.31 # ? 22.57.40 # title on the status bar? 22.57.46 # Pi 22.58.22 # is that what do you want to do? 22.58.31 # i do it allready 22.58.37 # did 22.59.24 Join akaidiot [0] (n=nope@c-7244e255.363-1-64736c11.cust.bredbandsbolaget.se) 22.59.58 # no, the code to pick the colors is abit big and so i want to get rid of all the bool functions 23.01.08 Quit NicoFR () 23.02.11 Quit _FireFly_ ("Leaving") 23.02.11 # has anyone with CVS commit looked at my RTC in WPS patch and have comments or suggestions or requests or anything? 23.02.51 # http://sourceforge.net/tracker/index.php?func=detail&atid=439120&group_id=44306&aid=1388592 23.09.04 # (H3xx, cygwin) hey, just did a CVS update and I get a build error: vorbis.elf section .iram is not within region PLUGIN_IRAM 23.10.16 Quit Hansmaulwurf ("( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de )") 23.12.15 # and i'm out, later all 23.12.37 # mm thereis a way to freeze the plugin until some command is complete? 23.18.31 # petur: Lear / my last chunk of optimizations grew iram use quite a bit, but I know he builds on cygwin as well... what compiler? 23.18.46 # 4.0.2 23.19.04 # gcc4 apparently makes bigger code than 3 still 23.19.11 # :( 23.19.23 # In my experience it also made slower code for the iRiver 23.19.31 # (based on profiling) 23.19.44 # gcc4 is not a good choice for rockbox 23.20.36 # has worked for me so far, I probably should rebuild my dev environment :( 23.21.01 # it _works_ it just isn't as good... 23.21.32 # s/_works_/_worked_ 23.21.47 # I have 4, 3.4 and 3.3 all built for comparison testing and 3.4 was the fastest in the tests I did a while ago (no big differences, but 4 was slowest) 23.23.32 # * amiconn should try to build m68k-elf-gcc 3.4.5 23.23.54 # * petur joins amiconn 23.24.27 # I used m68k-elf-gcc 3.4.4 so far 23.25.33 # This is recommended for coldfire, as is 3.3.6 for SH1. Updating to a newer revision was usually a good idea in the past 23.25.51 Join RotAtoR [0] (n=e@12-210-82-91.client.insightBB.com) 23.29.39 Quit akaidiot ("( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de )") 23.31.02 Join ataxy [0] (n=chatzill@modemcable189.249-203-24.mc.videotron.ca) 23.31.33 # 345 is my flavor. 23.33.25 # Ok. Let's see whether the cygwin fix is still needed... 23.33.38 # But first, binutils 2.16.1 23.35.26 Quit JoeBorn ("open.neurostechnology.com") 23.36.46 Join miner49er [0] (n=miner49e@82-32-84-226.cable.ubr01.chap.blueyonder.co.uk) 23.37.41 # hi there, what the play/pause button #defined as on the H120? 23.38.43 # isnt it BUTTON_START? 23.38.54 # or something like that ^^ 23.39.00 # just look in buttons.h 23.39.18 # i don't know, that's why I asked! :-) cheers 23.39.31 # #define BUTTON_ON 0x0001 23.39.45 # i think this should be it :) 23.40.15 # where is button.h in the directory structure please? 23.40.37 # firmware/export 23.40.39 # button.h 23.40.54 # cheers Matze41 :-) 23.41.01 # ^^ np 23.41.40 # hhmm, not any comments...do you know which button #define is the joystick click down one? 23.41.56 # button select? 23.42.00 # yep 23.42.03 # should be that one 23.42.42 # oh right, cheers. It's annoying I have an iRiver h120, but it doesn't work so I can't test my program on it!! 23.43.56 # btw lostlogic: i corrected my patch for fm radio... now without whitespace removing and without compression :) evrything okay like this? https://sourceforge.net/tracker/index.php?func=detail&aid=1393844&group_id=44306&atid=439120 23.44.17 # what kind of program are you writing? 23.44.50 # can anyone help me choose buttons for the iPod please. I need a pause, thrust, left, right, fire and hyperspace? 23.45.38 # there arn't enough are there...damn 23.45.45 # an asteroids game, lame i know. 23.45.45 # maybe you should go find a keylayout :) just a picture or something like that 23.46.05 # so you can see the keys and how they're arranged 23.46.22 # it's slightly fun on my archos recorder 23.46.48 # isn't there an astrorocks game already submitted? 23.47.02 # yes 23.48.42 # Matze41: looks good to me (not that I can actually commit it) 23.48.56 # noo i dont want you to commit that 23.49.12 Join Mongey [0] (n=mongeyc@213-94-252-135.b-ras1.dbn.dublin.eircom.net) 23.49.22 # but i want to make evrything right so it's easy for others to use :) 23.49.32 # yeah, i did it, just trying to get it working on the other platforms! 23.50.03 # miner49er: ahh, cool 23.50.40 Quit Mongey (Client Quit) 23.50.44 *** Saving seen data "./dancer.seen" 23.50.48 # someones tried it on the iRiver, but they didn't let on who they were... 23.51.18 # miner49er: sry i dont get that sentence.. but if u need someone who tries it out on iriver just ask me (H320) 23.53.04 # Matze41: Well, someone commented on the patch (in sourceforge) but didn't say who they were so I couldn't discuss it with them. If you could download the patch and tell what needs changing for it, I would appreciate it, cheers :-) 23.53.34 # miner49er: okay just gimme some minutes 23.53.50 # what's the middle button on the iPod called in button.h? That would make a good fire button!