--- Log for 04.06.107 Server: niven.freenode.net Channel: #rockbox --- Nick: logbot_ Version: Dancer V4.16 Started: 2 days and 15 hours ago 00.00.28 Quit Rondom (Read error: 113 (No route to host)) 00.01.26 Join Rondom [0] (n=Rondom@p57A97D1A.dip.t-dialin.net) 00.03.19 Part TrueJournals 00.03.53 Quit petur ("zzz") 00.10.47 Join ducbian_ [0] (n=Nick@cpc2-cmbg3-0-0-cust33.cmbg.cable.ntl.com) 00.18.49 Join Rob222241 [0] (n=Miranda@p54B159D5.dip.t-dialin.net) 00.22.08 Join perldiver [0] (n=say@cpe-72-225-231-80.nyc.res.rr.com) 00.23.34 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net) 00.24.34 Part n1s 00.28.30 Quit ducbian (Read error: 110 (Connection timed out)) 00.30.27 Quit pearldiver (Read error: 110 (Connection timed out)) 00.30.33 Quit Rondom ("Ex-Chat") 00.32.09 Quit Buschel () 00.36.24 Quit Rob2222 (Read error: 110 (Connection timed out)) 00.38.22 # good night at all 00.38.26 Part TrueJournals 00.38.57 Quit DerPapst ("So Long And Thanks For All The Fish!") 00.39.40 Join lxx [0] (i=lexx@217.172.21.68) 00.40.03 # hi 00.40.30 # what the best prog for upload mp3`s to my ipod? 00.41.35 Quit Arathis ("Bye, bye") 00.43.25 # lxx: depends, which OS are you running? and what are you looking to do? sync, or just transfer? 00.45.04 Join Davide-NYC [0] (n=chatzill@user-12hdtj8.cable.mindspring.com) 00.45.38 Quit Davide-NYC (Client Quit) 00.47.14 # i use windows 00.47.24 *** Saving seen data "./dancer.seen" 00.47.25 # for transfer 00.49.03 # You want something different to simply copying them in windows explorer? 00.50.06 Quit ompaul ("bye") 00.50.55 # lxx: you could use foobar's copy +rename feature 00.51.29 # but i ise now official firware 00.51.43 # i need prof for official firmware 00.51.51 # prog 00.52.15 # lxx: you want to run a dual boot iPod? 00.52.31 # maybe 00.53.57 Quit ender` (" I think I remember an episode of MacGyver where he overthrew a violent dictator with a rubber band, 2 bottle caps, and some ) 00.54.09 # lxx: This isn't a support forum for Apple's firmware... 00.54.21 # lxx: if you need dual boot, then any program that writes to the itunes DB will do (itunes, winamp, mediamonkey) 00.55.18 # where i can ask about id? 00.55.20 # ψε 00.55.22 # it 00.56.18 Part krush1704 00.56.46 # rockbox support fm transmitters? 00.57.32 # See this page for Rockbox's compatibilty with ipod accessories: http://www.rockbox.org/twiki/bin/view/Main/IpodAccessories 00.58.23 Join krush1704 [0] (n=andikrus@p57BB2208.dip0.t-ipconnect.de) 00.58.36 Join Rick| [0] (n=rhelmus@86.92.197.142) 00.59.02 Quit spiorf ("Read error: 110 (Connection timed out)") 00.59.41 # others fm tranmiters dont tested or just not working? 01.01.02 # Not tested. Or at least, not added to the wiki page. 01.01.21 Quit bluebrother ("leaving") 01.01.30 Join jhulst [0] (n=jhulst@c-71-205-0-132.hsd1.mi.comcast.net) 01.02.37 # but what the best fm transmitter? 01.02.55 # I wonder if anyone could help me with my ipod nano not wanting to (re)boot anymore? 01.05.39 # from http://commit-digest.org/issues/2007-06-03/ : "Support for RockBox-based devices in Amarok." anyone know about this ? 01.05.46 # Rick|: What's the problem? 01.05.51 Quit Entasis (Read error: 104 (Connection reset by peer)) 01.06.13 # well it seems that it happened after the battery was fully discharged 01.07.47 # I've recharged for a few hours, though everything seems to be dead; blank screen, ipod not recognized by pc, not possible to reboot (tried to toggling hold key) 01.08.01 # pretty much all fm transmitters are crap 01.08.13 # FM is not a very good medium for sending audio 01.08.23 # particularly at low power 01.08.48 # saratoga: agreed, aux input car decks are much better 01.08.59 # Rick|: How long have you been holding MENU+SELECT together? Some people need 30+ seconds. 01.09.31 Quit Siltaar (Remote closed the connection) 01.09.52 # but i whant find good transmitter with ipod charger and i dont have others options to connect ipod in car 01.10.04 # several times for about 10-20 and a few 30-40 01.10.12 Quit DC1 (Read error: 110 (Connection timed out)) 01.10.28 # guess it won't hurt to try it another time for about a minute 01.13.12 # nope 2 times ~1 min and still very sleepy 01.14.43 # I wonder if Rockbox could anything have to do with it? It went offline while booting RB 01.22.17 Part element_g ("Kopete 0.12.4 : http://kopete.kde.org") 01.23.11 Part krush1704 01.28.02 # so anyone here use fm transmitter in car? 01.29.15 # Rick|: not really...i would suggest that you leave it plugged in overnight, and then try it again 01.32.48 # scorche: ok...I'll try that and see if it's tomorrow any better 01.33.27 # Rick|: i've had this a few times before. what i've actually found is that it only seems to fix itself if you leave it *unplugged* for like a full day. 01.33.38 # :) 01.33.48 # perhaps as in, drain the battery... 01.34.22 # had the problem the first time, charged it all day on and off, no fix.. left it sitting for a day, it started working again. happened again a few days ago. left it unplugged for a day, started working again. 01.34.37 # than it might be longer, since it has been plugged in for a few hours already (assuming it actually charged) 01.34.48 # ok 01.34.59 # Rick|: well, i surely charged it "fully" all day before leaving it unplugged for a day. 01.35.03 Quit obo ("bye") 01.35.09 # ah cool :) 01.35.13 # still seemed to fix itself.. so *shrugs* 01.35.16 # that's my solution :) 01.35.37 # well I give it some rest then :) 01.35.47 # Rick|: that should help. :) 01.36.26 Quit Nico_P (Remote closed the connection) 01.36.36 # how did you found out it worked again? just plugged it in after a day? 01.36.41 Join BHSPitMonkey [0] (n=stephen@adsl-66-139-217-178.dsl.rcsntx.swbell.net) 01.37.49 # Rick: i left it sitting with the hold switch on for probably 18 hours, only tried it a few times.. and then i turned the switch off once at the end and pressed MENU, the screen flickered on for a few seconds before dying. 01.37.53 # the battery was completely dead. 01.38.20 # seems like you need to somehow let it discharge during that time before it'll come back on... 01.38.38 # i don't understand what causes it or why menu+select won't reset it, but it's happened twice and that's worked both times. quite annoying. 01.39.32 # hmm 01.39.40 Quit ducbian_ (Read error: 110 (Connection timed out)) 01.40.23 # did it happen when your battery was nearly drained aswell? 01.40.38 # i don't remember. but i definitely charged it fully before leaving it unplugged that period of time. 01.40.38 Quit Daishi ("Client exiting...") 01.40.44 # ok 01.40.48 # so it's like it's stuck on, draining the battery as if it were running... 01.40.53 # it wouldn't have died that quickly if it were "off". 01.40.58 # yeah 01.41.01 Quit XavierGr ("One firmware to rule them all!") 01.41.20 # oh well I hope it works :) 01.41.37 Part pixelma 01.41.40 # yeah, good luck. gotta run :) 01.41.42 # bbiab. 01.41.56 # ok thanks so far :) 01.45.05 Quit lxx () 01.45.38 Join webguest87 [0] (i=4a841a2a@gateway/web/cgi-irc/labb.contactor.se/x-005efe03dfcc8d1e) 01.46.19 # how can you guys see peoples ip adresses? 01.47.55 # helo again... 01.49.06 # grrrrrrrrrrrrrrrrrrrrrrrr... 01.49.49 # please dont spam the channel....the guidelines can be located in the topic 01.51.25 # Yes sir. Can i get my history removed? My brother got on whilst i was getting lunch. He is 13. And immature. Sorry... 01.51.49 # the logs are the logs, and they dont get edited 01.52.26 # but, for the purposes of this channel, i will not consider you as a spammer, unless it happens again 01.54.47 # I apoligize for the inconviences of my brothers doings. I swear not to do it again. 01.55.05 # well, if it was your brother, then you didnt do it before, no? ;) 01.55.23 # Exactly! 01.55.43 Part Rick| 01.57.05 # You mispelled hello during your first post... 01.57.17 # i know...i make a lot of typos 01.57.28 # so many, that i have stopped correcting them largely 01.57.35 # Ok. 01.58.13 # If no one is discussing anything, what do you guys do? 01.58.38 # well, this isnt a social channel, so it sits silent until needed 02.01.15 # OK. cool. I have a question. Where can I get the WAD files for Doom to work? I downloaded one and I cannot play any of my cool plugins? 02.01.36 # I hope they are free... 02.06.27 # Is anyone else on besides me and scorche? 02.08.58 # stupid question, how do i format a number with printf so that it gets zero padded if its small? 02.09.24 # I want to print with fixed widths for fp numbers 02.09.31 # "%+5.4lf" isn't doing it 02.10.18 # stupid question, why is there a setting that makes the screen display upside-down? 02.12.10 # oh never mind 02.12.58 Quit lids (Remote closed the connection) 02.15.59 Quit webguest87 ("CGI:IRC") 02.16.19 Join webguest87 [0] (i=4a841a2a@gateway/web/cgi-irc/labb.contactor.se/x-afb58b92a6dc88b4) 02.20.26 Join lids [0] (i=lds@gateway/tor/x-20e514624ee1eb89) 02.22.07 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net) 02.24.45 Quit webguest87 ("CGI:IRC (EOF)") 02.28.20 Quit entheh ("^~") 02.36.16 Join XiLe [0] (i=47ea5344@gateway/web/cgi-irc/labb.contactor.se/x-d48f32fa56e3ce88) 02.38.20 Quit XiLe (Client Quit) 02.40.10 Part TrueJournals 02.46.32 Join toffe82 [0] (n=chatzill@71.142.12.215) 02.47.20 Join krazykit [0] (n=kkit@ppp-70-229-30-254.dsl.ipltin.ameritech.net) 02.47.26 *** Saving seen data "./dancer.seen" 02.47.49 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net) 02.50.03 Join XiLe [0] (i=47ea5344@gateway/web/cgi-irc/labb.contactor.se/x-5dd3ea983086973d) 02.51.37 # are there any ways to buy music other than cd that i can play on rockbox? allofmp3 isnt accepting credit cards i guess 02.54.32 # emusic? 02.55.10 # iTunes, provided you buy hte tracks without DRM 02.55.36 # anything for the major record labels? 02.56.41 # i heard emusic is only independent stuff, and itunes is only emi records :( am i best off just buying the cd, quality aside? 02.57.12 Quit BigMac (Remote closed the connection) 02.58.02 # itunes is supposed to be getting most of the other records to go DRM-less eventually 02.58.03 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon) 02.58.12 # so i guess you could buy CDs for now 02.58.25 # mmm okay thanks 02.59.17 # you always got google...or are you looking for a more legal way? 02.59.32 # im looking legal 02.59.50 # ive been using google with intitle and such for too long 03.04.05 # haha 03.11.25 Quit XiLe ("CGI:IRC (EOF)") 03.17.09 Quit jhulst (Read error: 113 (No route to host)) 03.41.27 Join Trih`Xeem [0] (n=TrihXeem@bas4-toronto06-1279310944.dsl.bell.ca) 03.54.42 # what does Can't load from partition bad checksum mean? 03.55.06 # Well, I know it means that the checksum is bad, but what would be required to fix it? 03.59.31 Part TrueJournals 04.05.56 Quit sbeh (niven.freenode.net irc.freenode.net) 04.05.56 NSplit niven.freenode.net irc.freenode.net 04.06.22 NHeal niven.freenode.net irc.freenode.net 04.06.22 NJoin sbeh [0] (i=sbeh@serverstaff.de) 04.07.03 Quit Trih`Xeem (Read error: 110 (Connection timed out)) 04.32.54 Quit YouCeyE (Read error: 104 (Connection reset by peer)) 04.34.51 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye) 04.45.50 # has there been an ignore "the" patch added to the svn? 04.47.27 *** Saving seen data "./dancer.seen" 05.02.46 Join Brunellus [0] (n=luigi@unaffiliated/brunellus) 05.02.56 Part Brunellus ("Ex-Chat") 05.04.46 # alienbiker99: no, theres a lot of disagreement about how to implement that feature 05.05.09 # i think people dislike committing english specific patches, since rockbox supports many languages 05.06.41 Join secleinteer [0] (n=scl@70.230.165.253) 05.06.56 # ah ok, actually i think i saw one a while back supporting like german and other but oh well 05.17.45 Quit saratoga ("CGI:IRC (EOF)") 05.19.10 Join tradotto_ [0] (n=tradotto@237-71-14-216-arpa.cust.cinci.current.net) 05.20.29 Quit kkurbjun (Read error: 110 (Connection timed out)) 05.21.53 Quit tradotto_ (Remote closed the connection) 05.38.38 # apparently i've only been idle for 2 days 05.38.50 # i know i've been idle longer than that...what gives 05.39.11 # so, i have an interesting bug, about to searchf or it on the tracker 05.51.06 Join tradotto [0] (n=tradotto@237-71-14-216-arpa.cust.cinci.current.net) 05.51.17 # Hello 06.11.59 Quit rotator () 06.13.09 Join printfXh4 [0] (n=pseudo@ppp254-27.lns2.bne4.internode.on.net) 06.32.52 Join nick89 [0] (i=nick89@c220-237-70-197.kelvn1.qld.optusnet.com.au) 06.33.13 Quit nick89 (Client Quit) 06.39.00 Part toffe82 06.47.31 *** Saving seen data "./dancer.seen" 06.52.13 Join HellDragon [0] (i=JD@unaffiliated/helldragon) 07.01.02 Join NIXON-Trent [0] (n=euphoria@c211-28-46-16.frank1.vic.optusnet.com.au) 07.03.01 # hello 07.23.46 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au) 07.32.47 Join ducbian [0] (n=Nick@cpc2-cmbg3-0-0-cust33.cmbg.cable.ntl.com) 07.37.16 Join Brunellus [0] (n=luigi@unaffiliated/brunellus) 07.43.12 Quit tradotto (Read error: 104 (Connection reset by peer)) 07.45.12 Join _Veseliq_ [0] (n=veseliq@host.91-92-172-170.airbites.bg) 07.48.16 Join Rob2222 [0] (n=Miranda@p54b1696e.dip.t-dialin.net) 07.50.42 Part maffe 08.01.16 Join Siltaar [0] (n=Siltaar@193.52.208.229) 08.06.36 Quit Rob222241 (Read error: 110 (Connection timed out)) 08.08.33 Join pondlife [0] (n=Miranda@cpc1-rdng11-0-0-cust362.winn.cable.ntl.com) 08.10.17 Quit lids (Remote closed the connection) 08.11.07 # anyone here ? 08.16.25 # I am 08.19.35 Join Rob222241 [0] (n=Miranda@p54B16955.dip.t-dialin.net) 08.20.49 # ah u wouldnt know much about linuxipods movie playback ? 08.22.05 # NIXON-Trent, I know that it has been discussed a lot recently 08.22.39 # so it should'nt have already been implemented 08.22.58 # specially if the related user manual don't mention it 08.23.03 # do it ? 08.23.45 Join ender` [0] (i=krneki@84-255-206-8.static.dsl.t-2.net) 08.26.03 # i've been trying to find out if theirs can seek (fast forward and rewind) 08.26.20 # as its really annoying with the rockbox player as if u dont sit and watch the whole movie.. there is no point :/ 08.26.49 # as im often traveling on trains and so forth its annoying as i have to turn it off then back on and loose my place in the movie 08.28.34 Join LinusN [0] (i=linus@rockbox/developer/LinusN) 08.32.03 Quit Rob2222 (Read error: 110 (Connection timed out)) 08.35.37 # NIXON-Trent, I understand 08.36.00 # at least a cold resum function would be enough 08.36.48 # true 08.37.06 # so i was just seeing if the ipodlinux player supported some control options ... 08.37.31 # even though i do like rockbox better i'd just run linux on the side just for the player feature.. 08.38.42 Join lids [0] (i=lds@gateway/tor/x-9d722bde199b5976) 08.46.50 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-3a455a5c2e39d99e) 08.47.34 *** Saving seen data "./dancer.seen" 08.49.21 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb) 08.49.53 # NIXON-Trent: Which ipod do you have? 08.55.37 # amiconn: When you get time, please could you look at http://www.rockbox.org/tracker/task/7249 and see if this changes the .talk clip configuration sensibly? 08.58.03 # linuxstb: Btw, even an .rvf player will have the stuttering problem on ipods as long as the grayscale lib runs on core0 and no improvements regarding the lcd controller i/f can be made 08.58.36 # * GodEater is watching Linus Torvald's speed on source code management here : http://codicesoftware.blogspot.com/2007/05/linus-torvalds-on-git-and-scm.html : Quite interesting. 08.58.52 # Or rather, .rvf has its own "grayscale lib" mechanism, but the effect is the same 08.59.09 # pondlife: Will do... 08.59.15 # s/speed/speech 08.59.23 # (may need a reminder) 09.01.00 # amiconn: No problem.. It just seperates out the checking of .talk clips, allowing a fallback to the user's choice of spelling/numbers/none 09.01.00 # amiconn: Did you try that quick change to get the greyscale lib running on the COP? 09.01.53 # (s/CPU_INT/COP_INT/ in timer.c) 09.03.36 # linuxstb_ i've got photo ipod.. 09.03.37 # sorry was afk.. 09.03.56 Quit ducbian ("Leaving") 09.04.36 # NIXON-Trent: Then you could try this - http://ipodlinux.org/Mv_player (I've never used it though, so can't answer any questions about it - and they would be off-topic here anyway) 09.05.17 # But if you do get it working, I would be curious to know how the playback quality compares to mpegplayer. 09.05.51 Join petur [0] (i=d4efd6a6@rockbox/developer/petur) 09.08.45 Join austriancoder [0] (n=austrian@80.120.117.30) 09.09.00 # petur: ping 09.09.06 # * petur waves to austriancoder 09.09.40 # petur: so.. i have looked at each stack and i must say every one has a minus point - check wiki 09.09.59 # reading... 09.11.32 # WASABI-h is fine, because it has all from ubsotg to classes, but it could be harder to port it to other usb chips 09.12.00 # austriancoder: Do any of them rely on a malloc/free implementation? 09.13.15 Quit GodEater ("CGI:IRC (Ping timeout)") 09.13.40 # yeah thats what i was asking about linuxstb_ 09.13.52 # if it supported fastforward and rewind 09.14.03 # linuxstb_: WASABI-h... yes the others i am checking.. one moment... 09.14.14 # NIXON-Trent: It says it does on that page. 09.14.53 # austriancoder: That probably isn't a reason to reject a stack, but it's something you'll need to change. 09.15.05 # * petur is getting terrible work interrups :/ 09.17.10 # linuxstb_: I've been trying to port that wasabi stack to h300, the mallocs are used to allocate mem for the typical usb structures for each endpoint. That number should be limited so I created something with a fixed array of buffers 09.17.24 # not optimal but just trying to get some life in it 09.17.45 # petur: so.. should we use the wasabi stack? 09.18.26 # I really don't know. What do you feel about the code structure? 09.18.31 # LPC214x USB stack uses no free or malloc... usb4rt uses free and malloc 09.20.06 # Another question I would ask is if they all have active developer communities? It could be very helpful if there are devs around you could ask questions to. 09.20.45 # petur: the structre looks okay... but the interface to the hardware looks not so nice to me. Here i like the linux way much more.. but every stack has plus and minus points.. 09.21.22 # I had e-mail contact with the author of wasabi some months ago, but he's not responding any more. He did like the fact I was using his code 09.22.27 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-157a4ea2b6b5f3c3) 09.22.28 # austriancoder: yes, the hardware link is not so good, but then again it was only written for the isp 09.23.03 Join davina_ [0] (n=dave@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com) 09.23.58 # I'll try to look at that code too tonight.... damn real life and payed work 09.24.15 # petur: refactoring hardware interface and it should be okay... maybe we can take here some parts of usb4rt 09.24.36 Join B4gder [0] (n=daniel@static-213-115-255-230.sme.bredbandsbolaget.se) 09.24.40 # maybe.... have a look 09.25.41 # else I would start with a simple stack and use classes from wasabi 09.26.13 Quit Lynx_ (" HydraIRC -> http://www.hydrairc.com <- The future of IRC") 09.26.21 # that would be another option... but not much different from taking wasabi and changing the hardware interface 09.28.16 # also, I have no clue how good wsabi is rdevice-related, I've only been poking around in the host related stuff 09.28.37 # (fix the typos yourself) 09.30.00 # I have the same problem.. I have only looked at the source but I never seen it in action 09.33.24 # petur: Okay.. will use the wasabi stack for coding work. I will begin to import the stack in my local rockbox and I will spend some time to refactor the hardware interface.. is this a plan or not? *g* 09.34.23 Quit linuxstb_ ("CGI:IRC") 09.34.30 # petur: also you have done some work with it... so it would be no waste of time 09.34.37 # ok... but we do want to use the hardware part for h300 ... 09.35.17 # I'll zip up what I made of it but it's a real mess. I only wanted to get it compile asap and get life in it, so the structure is a bit borked 09.35.35 # hmmm.... remove that 'bit' at the end 09.35.46 # and the 'a' too 09.35.54 # ;) 09.36.47 # see if you can pinpoint the place where you can insert an interface to other hardware 09.38.04 # at some point, all usb drivers should be talking about endpoints and transfers, that would be the point to look for 09.39.24 # I've learned quite a bit about the isp1362, it's actually a nice chip. You give it a buffer and say: send this using that transfer type 09.39.47 # I don't know how other controllers do it, and certainly not the pp one 09.40.04 # sansas chip is a EHCI based one.. 09.40.35 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor) 09.41.40 # I'll have to read up a bit on USB controller standards, don't know how EHCI looks like 09.41.44 # will be back in about 2 hours... need to do some shopping 09.41.57 Nick austriancoder is now known as ac_away (n=austrian@80.120.117.30) 09.42.07 # ehci is quite similar to ohci ;-) 09.42.15 # ahci 09.42.20 # :) 09.42.26 # bless you 09.43.11 # ehci is a standard set of registers 09.43.20 # bsically 09.43.26 # what an extremely short wikipedia article about them 09.45.39 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma) 09.45.44 # The S3C2440 in the Gigabeat is OHCI compatible 09.49.30 # Is that accessible via the dock? 09.51.33 # http://www.intel.com/technology/usb/ehcispec.htm 09.51.46 # there's 155 page pdf there 09.51.46 Join miepchen^schlaf [0] (n=hihi@p54BF6F1A.dip.t-dialin.net) 09.52.39 # better print it in two copies :-) 09.52.50 # Did someone say... print!? :D 09.52.58 # bah... work stkov... 09.57.06 # LinusN: yes, it's is 09.57.14 # eh, that was for linuxstb 09.58.01 Join Lynx_ [0] (n=lynx@tina-10-4.genetik.uni-koeln.de) 09.59.44 Join Thundercloud [0] (n=thunderc@82.152.213.103) 10.02.27 # * linuxstb wonders if anyone has looked at http://www.rockbox.org/tracker/task/7256 10.03.50 # * JdGordon looked at the text... 10.04.51 # linuxstb: looks nice 10.05.23 Quit printfXh4 (Read error: 104 (Connection reset by peer)) 10.05.32 Join printfXh4 [0] (n=pseudo@ppp254-27.lns2.bne4.internode.on.net) 10.06.34 # isnt ape the codec which is supposed t be x8-centric? 10.06.38 # x86* 10.07.09 # dionoea: you don't have to prepend commit comments with the filename. in fact I'd like to ask you not to. 10.07.21 Quit Thundercloud (Remote closed the connection) 10.07.54 # JdGordon: what do you mean? 10.08.19 Join Thundercloud [0] (n=thunderc@82.152.213.103) 10.08.55 # JdGordon: Only so far as the bytestream consists of 32-bit little-endian integers (but the bytes are processed one at a time - so not in the same order they appear in the stream), and that the algorithm lends itself to MMX optimisations (vector math). 10.08.56 # its heavily optimized for x86 and wasnt intended for DAPs? or am I tihking of a different one? 10.09.08 # JdGordon: Yes, that's what's been written about it. 10.09.21 # alrighty, then good work getting that far :) 10.10.33 # * JdGordon getting very sick of this list code ;p 10.13.12 # JdGordon: Don't you have an exam? ;) 10.13.44 # yeah, friday... 10.14.02 # then a weeks break, then 2 the week after... but im not worried :'( 10.14.30 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean) 10.17.51 Quit GodEater ("CGI:IRC") 10.18.16 # are we allowd to use MAX() and MIN() macros? or best if we dont? 10.20.09 Join XavierGr [0] (n=xavier@ppp53-138.adsl.forthnet.gr) 10.22.15 # JdGordon: They're defined in system.h, and I can't think of a reason why you shouldn't use them. 10.22.31 # petur: Commiserations. I've had work stkov for about 6 months now... Good luck in popping. 10.22.41 Join Rondom [0] (n=Rondom@p57a96110.dip.t-dialin.net) 10.23.22 # linuxstb: ah cool, 10.24.06 # pondlife: they just shifted priorities, and the board I'm working on is now top prio :( 10.25.33 # or as my boss told me friday: get prepared to do some extra time 10.25.48 # extra time = extra pay? 10.25.54 # yes 10.26.02 # Not so bad then. 10.26.10 # but I'd rather have the extra time for myself 10.26.58 # * pondlife too 10.28.08 # B4gder: did you see that talk of Linus Torvalds about scm? http://codicesoftware.blogspot.com/2007/05/linus-torvalds-on-git-and-scm.html 10.28.27 # he made some famous quotes 10.28.36 # he frequently does :) 10.29.05 # cvs users belong in a mental institution, etc... 10.29.21 # * linuxstb checks himself in 10.30.16 # if I can take my laptop and they have good internet access, I'm in 10.30.32 # cvs ci self 10.30.33 # :p 10.30.38 # * JdGordon runs away 10.30.50 # Linus tends to get enthusiastic about his projects, which is generally a good thing. And he enjoys making pointed statements that fuel debate. 10.31.16 # He'd be called a troll if in fact he wasn't the boss :-) 10.31.25 # lol 10.31.29 # Did someone say... troll!? :D 10.32.58 # Linus trolling: http://realworldtech.com/forums/index.cfm?action=detail&id=79654&threadid=79612&roomid=2 10.34.56 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul) 10.35.38 # Galois: but still you can't deny that he has a very valid point 10.36.37 # he simply doesn't bother to try and be diplomatic 10.39.35 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-cca45102fea3eabf) 10.41.38 # I found most of the arguments he made for using git very compelling, if a little blunt :) 10.42.07 # Absolutely. In fact decentralized SCM like git would probably be a better option for rockbox than SVN. The problem is that it's a very difference concept and a very different tool, causing confusion and requiring training. So we chickened out and went for svn anyway. 10.42.22 # I was thinking the same thing 10.42.41 # esp. his points around "not checking in new code to cvs until it passes the test suite" 10.42.53 # clearly we don't have a test suite - but we do like patches that work :) 10.43.05 # yeah 10.43.06 # if git were being used then patches wouldn't need to be in flyspray 10.43.18 # they could exist in developer X's repo for the interested party 10.43.29 # and only get merged into the "master" repo when everyone liked them 10.43.42 # a problem with that is that many people do not have machines where they can publish their repos. 10.43.48 # so yo end up withh hundreds of forks/ 10.43.51 # a salient point 10.44.07 # JdGordon: yes, but git does make merging branches very easy 10.44.11 # JdGordon: yes, one fork per patch. are patches in text files really better? 10.44.26 # kernel.org hosts git repositories for hundreds of kernel developers 10.44.48 # linus said he does merges across 22,000 files 4 or 5 times a day 10.44.52 # Galois: yeah, we'd need something like that too 10.45.00 # in less than 5 seconds or so per merge 10.45.22 # Zagor: that's a lot of disk space! 10.45.30 # disk is cheap 10.45.39 # although granted we don't have *hundreds* of developers 10.45.58 Join Nico_P [0] (n=nicolas@jau31-3-82-239-20-145.fbx.proxad.net) 10.46.28 Quit miepchen^schlaf (Read error: 110 (Connection timed out)) 10.46.51 # I also liked what he said about the "commit access" politics 10.47.04 # could git and svn be run in parallel? 10.47.23 Join Rob2222 [0] (n=Miranda@p54B1787F.dip.t-dialin.net) 10.47.23 # JdGordon: sort of, but with more work than gain imho 10.47.35 # you could maintain a local git repository of what's in svn 10.47.37 *** Saving seen data "./dancer.seen" 10.47.42 # and use git for your local work 10.47.45 # yup 10.47.46 # GodEater: I think he got that wrong, you have worse politics, because people say fine dont take my patch ill fork and no contribute to you again 10.48.14 # JdGordon: that's a possibility certainly - he didn't say it got *rid* of politics :) 10.48.19 # JdGordon: generally that's not a problem for larger projects. developers want their code in the main branch. 10.48.39 # but we have a similar "problem" with flyspray 10.48.52 # people have patches in their they'd like to see commited 10.49.08 # and we don't all agree they should be in there 10.49.13 # access politics is a natural consequence of a centralized repository. of course anyone can commit when "commit" is local. 10.49.15 # I don't see that moving to git would change that 10.49.16 # yeah, but they dont have a dozen patches in one task, which effctivly would happen with git 10.49.36 # JdGordon: that would mean they weren't using git correctly 10.49.51 # Linus's ideal is that you have one branch per "feature" or patch 10.49.53 # ah, ok 10.50.24 # I dont think he expained that well then, or I tuned out during that part :p 10.50.25 # so part's of a developer's repo could be commited, whilst leaving behind something else that other's don't like as much 10.50.34 # it *was* a long talk :) 10.50.51 # I probably missed bits too by trying to read my email at the same time 10.50.55 # was still better than studing though 10.51.03 # hahaha - what isn't ? :) 10.52.51 # I think the mindset for a developer keeping *all* their changes in one branch is as a result of the difficulty you get with merging in CVS. When it becomes as easy as it is with git, it's more natural to create one branch per patch 10.53.08 # dinner :) bbl 10.53.10 # eek just tried to install a boot loader for ipod 10.53.16 # now ipod displays a folder with ! on it 10.53.19 # and the ipod url ... 10.53.26 # I'm not really in a position to comment though, I've not done a lot of work with either system 10.53.49 # GodEater: not entirely. it's also the feeling of making accumulating improvements 10.54.37 # Zagor: that's natural when the features are in a working state - but there'd be no point accumulating multiple "work in progress" features 10.54.47 # IMHO anyway 10.55.47 # anyone know a simple way to get linuxipod + rockbox + apple os on a boot loader screen ? 10.56.06 # Could SVN branches be set up for major upcoming features? I'm thinking the GSoC projects as an example... 10.56.07 # ipodlinux is offtopic here 10.56.24 # well anyone know how to get a boot loader 10.56.29 # that shows rockbox + apple 10.56.33 # and i can pick which to load ? 10.56.42 # If you installed Rockbox according to the installation instructions, you have that. 10.56.46 # ipl's bootloader, which is what you're talking about, is also offtopic 10.56.54 # linuxstb: I've been thinking about a system to promote more branches. But the access system is the biggest stumble. 10.57.06 # * GodEater votes we move to git :) 10.57.17 # * GodEater is fully aware his vote is worth nothing. 10.57.20 # :-) 10.57.21 # lol whats not off topic lmao 10.57.25 Join maffe [0] (n=maffe@195.159.148.248) 10.57.35 # linuxstb what do u mean if i installed rockbox ? 10.57.37 # NIXON-Trent: rockbox related topics, unsuprisingly 10.57.48 # Zagor: For SoC, 2 of the students already have commit access, and I can't see a problem with giving the other two students access - with "social" restrictions. 10.58.37 # NIXON-Trent: The Rockbox bootloader you installed lets you boot either Rockbox or the Apple firmware. You just need to make the decision before turning on your ipod (and then turn the hold switch on or not) 10.58.40 # * GodEater looks forward to hearing what these social restrictions are 10.58.52 Quit Kohlrabi (Remote closed the connection) 10.58.54 # linux any alternatives to that 10.58.56 # are we not allowed to talk tot hem ? :) 10.58.59 # agreed. my idea was a more general concept, to get some of the git benefits (branches instead of patches) 10.59.01 # incase i want to choose from 3 boot options ? 10.59.23 # NIXON-Trent: the rockbox bootloader will also boot iPL 10.59.31 # you hold play when booting 10.59.35 # IIRC 10.59.54 # NIXON-Trent: If you install IPL according to their instructions, you'll install their bootloader, which will also detect Rockbox and give you an option to load that. 11.00.04 # IPL ? 11.00.11 # iPodLinux 11.00.27 # http://www.ipodlinux.org/Loader_2#How_it_works 11.00.31 # i was trying to install that 11.00.39 # and well it seemed to bugger my ipod and i just had to restore it then :/ 11.00.44 # lolz 11.00.53 # <3's to rockbox being way easyer :P 11.01.08 # only reason i want ipodlinux is because u can rewind and fast forward movies :/ 11.01.46 Quit GodEater ("CGI:IRC") 11.01.59 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-9cf2cd089f147f00) 11.03.09 Quit Rob222241 (Read error: 110 (Connection timed out)) 11.03.46 Quit lids (Remote closed the connection) 11.04.19 Join Kohlrabi [0] (n=Kohlrabi@frustrum.nosebud.de) 11.04.57 Quit jhMikeS (Nick collision from services.) 11.05.03 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS) 11.05.14 Join lids [0] (i=lds@gateway/tor/x-abad39361930f3b0) 11.06.57 Quit idnar (Nick collision from services.) 11.06.59 Join idnar_ [0] (n=mithrand@unaffiliated/idnar) 11.11.12 Quit aliask ("ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007060115]") 11.11.23 Quit SaLoMoN (Read error: 110 (Connection timed out)) 11.12.35 Join SaLoMoN [0] (n=SaLo@xglusers.de) 11.19.00 Join obo [0] (n=obo@host217-41-62-170.in-addr.btopenworld.com) 11.20.09 Join desowin [0] (n=desowin@avc146.internetdsl.tpnet.pl) 11.21.00 Quit ze (Read error: 104 (Connection reset by peer)) 11.21.12 Join ze [0] (i=ze@cpe-76-175-22-254.socal.res.rr.com) 11.23.56 Quit desowin (Client Quit) 11.24.28 Join spiorf [0] (n=spiorf@87.13.120.224) 11.28.16 # anyone interested in testing out my new list code which I'm pretty sure fixes all known bugs because of this silly accelleration stuff? 11.29.18 # Were you able to reproduce the Sansa problem with slow scrolling? 11.30.03 # yeah, the problem is the button dirver doesnt send button_rel events which is confusing te action code 11.30.42 # Did you manage to fix that, or is your patch for other bugs? 11.30.51 # the easiest fix is to check how long since the last action, but havnt got that working yet 11.31.20 # so what is your current fix then? 11.32.25 Nick ac_away is now known as austriancoder (n=austrian@80.120.117.30) 11.33.15 # FS#7258 11.33.20 Join Rick| [0] (n=rhelmus@86.92.197.142) 11.33.45 # makes the selection code more similar to how it used to be, moving up/down the list works nicely again 11.33.46 Join ljcrabs2 [0] (n=alien5@124-197-24-173.callplus.net.nz) 11.35.50 # ok, so this doesn't really solve the acceleration problems, but merely the issues with the list cursor? 11.36.12 # midkay: In case you are here; I wanted to thank you because my iPod is now working again :) 11.36.14 # yep 11.36.18 # i see 11.36.51 # which accell problems? I thought the main problem was the bad language for the settings? other than that it works doesnt ti (except that wierd sansa bug) 11.38.31 # petur: I am waiting for your mail 11.38.32 Join Entasis [0] (n=Jarred@ppp7-94.lns10.adl2.internode.on.net) 11.38.45 # JdGordon: I think the problem is on the wheel targets - the acceleration needs to take account of the speed, but it doesn't. I like how it works on button targets though. 11.39.26 # austriancoder: it will be tonight, don't have it with me.... and lots of work here :/ 11.39.35 # JdGordon: Should this patch also fix the Iriver LCD remote problem? 11.39.41 # petur: no problem 11.40.26 # pondlife: hopefully, i've been testing in the h300 sim,and found no problems 11.40.34 # god ipodlinux is so hard to install things on 11.40.36 # grrr 11.40.48 # JdGordon: Urgh, something is horrid with SVN+the patch though... 11.41.13 # It's hard to describe but I'll try and put a note on Flyspray 11.41.49 # NIXON-Trent: again, iPL is not what we want to discuss here. Please take it to #ipodlinux. 11.41.50 # also, I tried reporoducing your follow playlist bug (before these changes) and couldnt 11.42.12 # I don't think that's list related - it's buffering related 11.42.28 # With your patch it doesn't allow the cursor bar to go to the top or bottom of the screen before scrolling. 11.43.14 # ? 11.43.38 # I'll try and explain with pictures on Flyspray... 11.44.08 # cool 11.50.35 # JdGordon: OK, look at http://www.rockbox.org/tracker/task/7258 11.52.48 # Let me know if that makes sense. 11.52.58 # it does with the pics :) 11.53.02 # but I thought I fixed that 11.53.18 # Can happen with both UP and DOWN. 11.53.50 # defiantly works correctly here... 11.54.10 # Do you want my config? 11.54.23 # oh wait... hmm.. 11.54.30 # it happens when you swap directions 11.54.34 # Yes 11.54.52 # And there are undisplayed items in the direction of travel. 11.55.51 # this one is going to be fun to fix :'( 11.56.26 # Is it me, or has the list code got over-complex recently? 11.56.27 # ;) 11.56.38 Quit Bagder (Read error: 110 (Connection timed out)) 11.56.43 # Can it not be KISSed back to health? 11.56.45 # no, its always been bad 11.56.50 # Ah, ok 11.56.54 # thats what this was doing... 11.57.03 # Good, KISS it better then!! 11.58.30 # does anyone else think it would be a good/(fun?) idea to start the apps/ from scratch? 11.58.33 # Zagor: i am only 20-min into the talk, but how does distribution work with telling people to download "our binary"?....if it is answered later, feel free to tell me "WTFV" 11.58.57 # JdGordon: Yep, time for Rockbox NT :/ 11.59.31 # scorche: the idea is that people grab whichever they want... 12.00.13 # sounds like a headache to support 12.00.33 Part maffe 12.00.40 # pondlife: im semi serious... start apps from scratch so we dont have to keep dealin with 3+ year old code.... 12.00.43 Join maffe [0] (n=maffe@195.159.148.248) 12.00.45 # linuxstb: just wanted to add a correction to your post - but then I saw you already edited it ;) 12.01.00 # JdGordon: 3 year old *debugged* code 12.01.05 # well yeah 12.01.24 # but i understand the feeling 12.01.27 # On the ipod with a dock connected hardware, is it the ipod software or hardware that handles commands, eg mute/next track etc? 12.01.31 # refactoring isn't all bad 12.01.40 # but its 3 yr old code + multiple hacks for each feature... 12.01.47 # yes 12.01.51 # scorche: "our binary" would be one build from the "master" repo at rockbox.org 12.02.07 # much as the "master" linux kernel tree is kept at kernel.org 12.02.37 # JdGordon: we could/should rewrite large part of rockbox, and we often do 12.02.38 Join rift [0] (n=heroes@25.56.70-86.rev.gaoland.net) 12.02.41 # Could we possibly start a experimental branch for this? 12.02.44 # GodEater: and this "master" would still have to have multiple guardians to say "we want this part and not this", no? 12.02.53 # pixelma: Yes, I'm always discovering new things Rockbox can do... 12.02.54 # like the action concept, or the settings code 12.03.09 # two examples of recent refactoring 12.03.10 # scorche: yes it would 12.03.13 # LinusN: but that even causes other problems, refactoring still depends on other hacks which we just cant get rid of anymore 12.03.42 # JdGordon: you should read "The UNIX Philosophy" 12.03.43 # scorche: well - in theory it could be one - but in practise I suspect we'd still end up with our "core" devs deciding amongst themselves what to pull in 12.04.56 # we won't be able to rewrite rockbox without buildind a monstrous "second system of man" 12.05.29 # hmmm...i could see where it would simplify branches and big changes, but other things, it would just add extra responsibilities onto people, and small changes would be simpler just with posting patches 12.05.56 Quit B4gder ("It is time to say MOOO") 12.05.57 # * scorche resolves to shut up and make comments after viewing the entire vid 12.06.05 # * LinusN goes to lunch 12.06.39 # scorche: the benefits of git would be felt most by the devs themselves, I accept that it would possibly make life more complicated in the short term for the master build system maintainer 12.06.58 # linuxstb: are you planing on doing "rockbox independant" releases of your monkey audio decoding lib ? 12.07.13 # dionoea: Probably, yes. 12.07.28 # GodEater: i realize that, but...hmmm....i will talk to you about it in 40 min 12.07.30 # cool :) 12.08.06 # scorche: heh - no problem. It's very academic at the moment given the change to svn only 6 months ago :) 12.08.33 # wouldnt that make it a better time to convert then later on when we are even more entrenched in svn? 12.08.40 # :D 12.09.19 # dionoea: Although assuming no-one objects, I would like to use Rockbox SVN as the "master" repository for it, so I don't have to worry about maintaining multiple versions. But I need to think of a way to unobtrusively add the standalone parts to Rockbox's SVN (it's only a few small files). 12.10.27 # linuxstb: I'm sure that once media players begin using it you'll get ASM patches to optimize it. I don't know if that would be ok in the rockbox source (It would be ifdefed out easily) 12.11.57 # dionoea: I still need to decide on a final license for it. Currently it's GPL'd (mainly because I've taken some code from another GPL project), but I'm aware that may cause problems for other projects. 12.13.08 # it wouldn't for VLC :) 12.13.28 # I don't know about others, but in the end you're already working loaclly in a git way, having multiple checked out working copies to do multiple things 12.13.51 # it does seem like it would simplify many things 12.14.23 # dionoea: ffmpeg is the main project I'm thinking about. It would be nice to get the benefit of the skills of their developers. 12.14.41 # They have GPL and LGPL parts ... but they don't like external libs. 12.14.42 # but you would need a maintainer that pulls changes into the master, unlike now where every dev with commit access pushes it into the repo 12.14.52 # They always try to write their own codecs 12.14.52 Quit lids (Remote closed the connection) 12.15.05 # and make collaborating much easier as well, but there are still a few things i am leary about (which i can address after viewing this in its entirety) 12.15.22 # yeah...it seems to put a lot more load on the maintainer(s) of the head 12.15.51 # scorche: yes, it's the one thing I didn't spot in the talk - how that part is managed. 12.16.00 # dionoea: Do you know about any other open source media players? Would they be happy with a GPL'd lib? 12.16.09 # scorche: although the kernel doesn't suffer from having to distrubute binaries like we do 12.16.21 # scorche: which makes life a hell of a lot easier 12.16.27 # indeed 12.16.31 # linuxstb: MPlayer and Xine would. 12.16.42 # dionoea: My feeling is that proprietory software can just use the official decoder (which I assume they all do already). It's mainly gpl'd software that has problems with Monkey's. 12.16.53 # exactly 12.18.26 # scorche: I think the issue here really is that Linus is viewed as the "god" of kernel development. He maintains his own repo, and that's what most people trust. With RB we don't have someone who's the "father of dev", so we don't have a key repo to trust for use as the "master". 12.18.31 # overall, at this moment, my opinions are: For collaboration between devs and for experimental branches/projects, it is great and a huge improvement; It comes at a huge cost of putting a ton of load on the maintainer(s) 12.18.44 Quit GodEater ("CGI:IRC") 12.18.57 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-7f286a19bdfebb88) 12.18.57 # GodEater: and he is perfectly willing and has tons of time to be that master 12.19.11 # scorche: well, he gets paid for it :) 12.20.09 # I don't know if this is because we don't have branches in the code, but I've noticed that it's quite rare for Rockbox devs to collaborate on new features. It's mainly people working alone - which isn't always the most productive. So a more flexible scm could help that. 12.20.27 # for git, it expends more time to say "hey , check this out please" "alright...hold on" "cool...i am pulling lines xxx in...should i pull xxxx as well?" 12.21.22 # basically, it seems like it comes down to "Is the flexibility and all that really worth all this extra load that a few get to carry?" 12.22.10 # as a developer, it seems obvious which choice to go with, but if one is to think of others... 12.22.15 # unless git allows a group of maitainers on the master, so we would still push changes instead of having them pulled like linus keeps saying 12.23.09 # im sure it does (at least i hope so), but it still adds some extra effort for some 12.23.22 # I think the real question is, do we currently have a problem that will be solved by better SCM? 12.23.48 # Will it get some of these patches into the repository more quickly than currently, and will it allow the better sort of collaboration linuxstb was talking about or will we still have the same sort of lone-wolf working going on? 12.23.58 # well, i wouldnt say it is a problem, but git simplifies many things and allows much more room for better collaboration 12.24.28 # at least it seems to, at this point 12.24.53 # But is SVN getting in the way of collaboration currently? It seems like most people currently tend toward working on a feature themselves, and the question is if that's because it's hard to collaborate, or just how the people around here tend to do things? 12.25.03 # I think for the sort of work we see our core devs do, where they submit patches to flyspray, git makes more sense 12.25.05 # * scorche wonders if we can pull linusT in here and have a little talk 12.25.20 # * JdGordon suspects its more than just svn that stops the sort of collaboration that would be nice... 12.25.55 # SVN does for a bit...it is much harder and more effort to have to apply patches and such when working on each project 12.28.17 # it makes things much simpler and faster to ignore the patches and just pull things here and there 12.28.26 # I feel git is better for big projects, and rockbox isn't that big so the benefits would not be much 12.29.09 Join arkascha [0] (n=arkascha@mailout.imageware.de) 12.29.21 # in terms of the benefit with collaboration, i feel it is, but as for the duty of the maintainers, i am not sure if it is big enough as well 12.29.30 Part arkascha ("Konversation terminated!") 12.29.34 # scorche: you would still need a place where people can advertize their 'branches' and others to comment on. I think I like flyspray much 12.31.14 # petur: this is very true as well 12.31.20 # i like it a lot as well, but the whole "here is a patch of some part of xxxxxunfinishedwork" can be improved 12.31.41 # I think one of the major problems of multiple GIT branches will be that people will probably just use them for unsupported builds. 12.31.45 # Also, would git retain the history of the patch when it was merged? 12.31.55 # You'll have 50 branches that all have 20 different modifications, all nearly identical 12.32.10 # Unless it's possible to easily prevent that. 12.32.25 # * petur says: lunch! 12.32.28 Join norbusan [0] (n=norbusan@dhcp06.algebra.tuwien.ac.at) 12.32.36 # There will be a lot of non-development cruft trying to float in, I think, unless it's easy to actively prevent this. 12.33.01 # Llorean: yes, but i am mainly thinking about current committers and people who write patches often 12.33.04 # Llorean: yeah, thats what I thought also, but you'd have an official set of branches, and then the rest hostig their own 12.33.15 # Llorean: I think the intention would be one branch for each patch. So if anyone tried to use them for unofficial builds, they would be breaking that rule. 12.33.47 # Llorean: we would be relying on a head branch for binaries and such 12.33.51 # linuxstb: Yeah, then someone has to actively keep track of the rule-breaking ones and do whatever's necessary to remove them. 12.33.57 # you'd still have the problem where patches/branches fall out of sync with the master wouldnt you? git doesnt magixcally fix that? 12.34.03 Part norbusan 12.34.08 # Llorean: not really, if i am understanding this right 12.34.14 # Llorean: That's the same as flyspray's "one feature per task" rule. 12.34.41 # And that still doesn't always get enforced. 12.34.49 # JdGordon: yes, but it would be much easier to re-sync things, i would think 12.35.01 # usually only on tassks which are not going t be commited anyway, i.e AA 12.35.23 # What happens when person A abandons a task, and person B picks it up? 12.35.30 # With Git do both have separate repositories? 12.35.34 # scorche: how could it? if 1 line changes, the same line would have to be fixed by a maintainr regarless of the scm 12.35.41 # Llorean: yes they would 12.35.46 # GodEater: I don't think I like that 12.35.51 # everyone has their own copy of the reop 12.35.53 # *repo 12.35.54 # With Flyspray you just post it on the end of the existing task. 12.36.05 # so ? 12.36.13 # So if Patch X gets worked on by seven people, you can see where it ends up. 12.36.28 # yep, massivly massive amonuts of forks 12.36.29 # Llorean: but person B has pulled in person A's work 12.36.39 # GodEater: Assuming you know about person B. 12.36.39 # Llorean: complete with history 12.36.54 # And assuming Person C has noticed and hasn't done some of his own work, with his own fork of the task. 12.36.55 # JdGordon: because you dont have to work with patches and pull it from their source itself 12.37.04 # well if we're still talking about using flyspray, instead of posting a patch 12.37.08 # you post a link to your repo 12.37.09 # The presentation of Flyspray is more linear, which I think just makes it clearer to anyone picking up that it's one task. 12.37.19 Quit jhMikeS (Nick collision from services.) 12.37.21 # I fail to see how this would stop people using Flyspray 12.37.23 # Multiple repositories I think could lead to someone missing that person B took over, and restarting from Person A's work 12.37.25 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS) 12.37.37 # you're just not posting files anymore 12.37.43 # Well then what's the real benefit? 12.37.50 # Llorean: which is why this is all about collaboration which includes communication 12.37.53 # the benefit is to the DEVELOPER 12.38.05 # Not "to whom is the benefit" 12.38.09 # "What is the real benefit" 12.38.50 # does anyone know how to get a diff of the whole tree btween two revisions? (in svn of couse :p ) 12.39.32 # GodEater: I understand pulling to the main repository is easier than all this work with patches. 12.39.44 # Llorean: the benefit I see is to the dev. let me explain : 12.39.57 # Let's say your developer A, you're working on feature X. 12.40.12 # developer B posts a patch to supprot feature Y, and wants you to test it. 12.40.15 # * scorche deletes his response to see where this is going 12.40.29 # with svn, you've got only your one repository to manage easily 12.40.45 # which already contains "polluted" code from your feature A. 12.40.54 # you dont' really want to add more stuff at this time 12.41.17 # so you'd have to check out a whole new repo, and patch in feature Y to make sure it doesn't mess with your work on feature X 12.41.30 # Llorean: honestly, i am not really confident in my ability to explain it properly...i would recommend watching the video of the talk 12.41.31 # in order to statisfy developer B's request for testing 12.41.52 # with git, you've already taken out the full "master" repo 12.41.57 # youve' branched it for feature X 12.42.21 # and now to satisfy developer B's request, you simply branch again off of "master" to include feature Y 12.42.29 # this hasn't messed with your original work at all 12.42.30 # Ah 12.42.34 # and you've still only got one repo 12.42.45 # now you find feature Y is working great. 12.42.55 # you merge it back into your copy of master, AND your feature X branch 12.42.58 # and continue working 12.43.06 # all with a minimum off fuss 12.43.12 # Llorean: http://www.youtube.com/watch?v=4XpnKHJAok8 12.43.15 # See, I just have a small script that takes a diff of my current sources, saves it, and reverts, and another that reverts and reapplies that diff, making it easy to test an individual patch and restore from the 'bookmark' with a minimum of fuss. 12.43.53 # Llorean: but that won't let you merge both features in until they both get commited to the master tree 12.44.10 # and you check them out one revision at a time 12.44.37 # I'm not so sure why I'd want to keep the feature around unless it was going to be committed anyway. 12.44.45 # my case there is extremely over simplified too - it only deals with two patches at once 12.44.46 # does anyone know what to do about a lang string which I want to deperacte but expect to reuse it later? 12.44.59 # hmmm...he describes a bit about using both SVN and git at the same time in somewhere around 37 mins remaining 12.45.10 # Llorean: perhaps you'd like to WORK on the feature? :) 12.45.20 # GodEater: Then I reapply my diff without reverting first. 12.45.31 # scorche: that's the unofficial use of git in organisations bit I assume ? :) 12.45.38 # yes 12.45.41 # Llorean: see, thats the svn way of thinking... 12.45.54 # Llorean: yes, but then two days in you think, actually, no I didn't want it after all. No you're in a mess. 12.46.25 # GodEater: And if their diff won't remove cleanly, why would the git branch remove cleanly? 12.46.31 Join Kanashimi [0] (n=devnull@scalehaven.com) 12.46.36 # it's not being removed 12.46.46 # you've got it all on it's own in it's own branch 12.46.58 # JdGordon: If you want to reuse it later, then it's not deprecated... 12.47.00 # if you've done it right 12.47.15 # GodEater: Ah, I thought you were talking about incorporating it into the code you're currently working on for some reason. 12.47.29 # linuxstb: So just change the string to ""? so it doesnt waste bytes? 12.47.33 # Llorean: you might have done, but you've branched the code you were working on first :) 12.47.39 *** Saving seen data "./dancer.seen" 12.47.41 # GodEater: All it sounds like is a way to manage .diffs 12.47.52 # Llorean: that's all any SCM is :) 12.47.55 # JdGordon: I would just keep it as it is. 12.47.55 # Yeah 12.48.05 # But it sounds like a way to manage diffs that makes other things more complicated 12.48.16 Join DerPapst [0] (n=DerPapst@pD9EB3C13.dip0.t-ipconnect.de) 12.48.17 # Llorean: I don't pretend to be an expert by any means, these are just the advantages *I* see. 12.48.19 # Rather than simply having a folder of diffs and applying the one you want to work on before you start, and taking a diff when you're done working 12.48.46 # perhaps to you - I don't find it that complex, and I personally don't like your chosen solution. 12.48.53 # but we're allowed to disagree :) 12.48.55 # Aye 12.48.57 Join safetydan [0] (n=dan@rockbox/developer/safetydan) 12.49.09 # Llorean: (really, watch the vid....i am trying to not comment much until i do, so as to understand it more) 12.49.19 # yeah definitely watch the video. 12.49.26 # git would be good for long running development patches like the lang cleanup 12.49.26 # JdGordon: in case you haven't looked it up already: svn -r diff 13337:13456 > blah.diff 12.49.28 # the google people he's talking to ask smart question :) 12.49.37 # +s 12.49.39 # DerPapst: thanks, found it :) 12.50.11 Join lids [0] (i=lds@gateway/tor/x-34524e49d08945d8) 12.50.15 # Hello there. I hope no one minds me asking a non-rockbox specific question here. Does anyone here happen to own a Cowon A2 player? I've been hearing it has trouble with MP3 files encoded with LAME -new VBR. I ordered mine a bit ago and just found out about the issues, so I'm a bit worried. How bad is the glitchyness? I don't necessarily need perfect audio for the moment, 12.50.20 # but hopefully it's not too terrible? 12.50.35 # Kanashimi: we do mind =) 12.50.59 # * GodEater runs to get a sandwich 12.51.18 # Kanashimi: Your question is off-topic, but I also don't think anyone here will know the answer - the A2 isn't supported by Rockbox, so owners of it don't hang out here. 12.51.45 Quit Brunellus (Read error: 110 (Connection timed out)) 12.52.51 # Alright, sorry about the bother :) Been using Rockbox on a iRiver of mine for a while and have visited before too. Just kinda worried and was hoping I might find someone by chance. 12.54.30 Join norbusan [0] (n=norbusan@dhcp06.algebra.tuwien.ac.at) 12.56.13 Part norbusan 12.56.19 # besides, it supports ogg, so why bother with mp3 12.57.08 # unless you happen to have some mp3 files for which you do not have a lossless copy (hmm, wonder how that could happen) 12.57.59 # maybe some users don't like ogg? 12.58.17 # pretty pointless discussion 12.58.18 # There's also online music stores, that aren't even in gray areas, that sell MP3s these days 12.58.39 # Or the CD being lost/damaged/stolen, ... 12.59.02 # linuxstb: then you have to destroy the MP3 files as well of course :) 12.59.02 # well the vast majority of legitimate online sales are in AAC format, and I don't see Cowon users complaining about lack of AAC support 12.59.28 # Galois: So all MP3 users are thieves? ;) 12.59.40 # Galois: what kind of complaints do they have? 13.00.01 # Galois: Don't forget they're DRM wrapped AAC, so you can't legally use them in the US even on an AAC supporting player 13.00.47 # I mean, even the non-DRM files offered for purchase are in AAC format 13.00.55 # Heh, let's not start on that kind of tangents. Well, there are a couple of things. While I do like ogg as a format it does also seem to require more processing which reduces battery life. And it is after all still the most common music format out there, used by independent artists and what not. 13.01.45 # Anyone looked at git-svn? Might be a nice halfway house for those who want git without forcing everyone to use it. 13.01.46 # does --preset invoke --vbr-new? 13.01.50 # Galois: The majority are DRM wrapped AAC. Then DRM Wrapped WMA from the Playsforsure stores, and MP3 in most of the non-DRM stores I've seen. Which store did you have in mind that was both a majority of online music sales, and DRM-free AAC? 13.02.26 # * DerPapst just got a comercial for an USB turntable ^^ 13.02.42 # sorry, I wasn't very clear: all files sold by the leading vendor of legal online music sales, whether DRM'd or not, are in AAC format 13.02.50 # Yes 13.02.58 # safetydan: did a pull last night - setup a few branches, seemed to work okay 13.03.00 # But until you can support the DRM, lacking AAC support doesn't matter 13.03.07 # So expecting complaints about it is silly 13.03.15 # Galois: while looking for music under the creative commons, all music came in mp3 and not all of them provided the music lossless as well 13.03.25 # safetydan: i was thinking something like how he said it in the answer to a question, but i really think it would be better discussed when people have done more research on all this, so we can speak from a level platform 13.03.27 # obo, safetydan: What would "git-svn" provide the typical Rockbox dev? 13.03.44 # linuxstb: local branches 13.03.58 # linuxstb: a local git repository, with all git's features, that's easy to keep in sync with svn 13.04.08 # linuxstb: provided you keep proper track of your own branches :) 13.04.21 # I believe it's what dana uses 13.04.32 # DerPapst: I've seen a guy DJing at a party with a USB turntable, worked very well 13.04.32 # scorche: true... I've just been looking at SCM systems for work related purposes lately as well 13.04.56 # Would someone like to write a wiki page describing how git-svn could be used with Rockbox? 13.05.08 # Sounds useful to me - I would source control in my local branches... 13.05.42 # dan_a perhaps rather. 13.05.50 # i wouldnt mind, provided i had a bit of help (after i get done researching and viewing all this) 13.06.05 # http://git.or.cz/course/svn.html <- me skimming... 13.06.08 # git also provides other more esoteric features 13.06.10 # Well I've only just typed "apt-get install git-svn" so... 13.06.15 # * scorche volunteers GodEater to help him 13.06.17 # Does git-svn still focus on the mentality that there's one key branch, and nobody really gets to commit to it, instead it has to pull? 13.06.18 # markun: i've never seen such a thing... but it's cool though. i mean there are washing machines that can use internet.. how usefull is that?! 13.06.24 # scorche: happy to 13.06.55 # * scorche wonders when Llorean will have a chance to watch the vid =P 13.06.57 # git-svn just pulls a central svn repo into a local git repo 13.07.05 # scorche: I'm watching it. 13.07.13 # Llorean: git-svn has no impact outside of an individual developers machine. It just means I can track my change sets locally before pushing it to svn. 13.07.36 # although git-svn can also commit back to SVN 13.07.41 # this is true 13.07.45 # safetydan: Ah, so it's still a commit based system, but locally you're effectively using git? 13.07.50 # Llorean: yup 13.07.53 # yep 13.09.14 # git-svn would mean no changes necessary at the rockbox server end - it still just runs svn 13.09.55 # DerPapst: don't know if we were talking about the same thing: http://news.harmony-central.com/Newp/2003/DJ-Console.html 13.10.16 # git-svn also seems to be a way to lessen the load on what would be put on the head maintainer(s) under using pure git 13.10.37 # git-svn still requires commit access 13.10.47 # Llorean: though because you're using git, it's possible for developers to share git repos while developing patches and still keep track of who did what. 13.11.24 # i do like how git tracks "content" instead of "individual files" 13.11.28 # safetydan: That sounds better to me. 13.12.06 # although the keeping track of who did what is only maintained in those dev's git repo's 13.12.19 # markun: i meant something like this: http://www.firebox.com/product/1401?src_t=cat&src_id=audio 13.12.20 # as far as the central repo is concerned, it was all done by whoever finally commits it 13.12.21 # I don't really like getting rid of the idea of commit access, because then you're depending on the central maintainer to do all the work of pulling. 13.12.36 # GodEater: yeah...that would be one strong limitation of it 13.12.47 # Llorean: it's still automatable - it would just take work 13.12.53 # It makes sense when you're doing a lot of non-live work, among a lot of people. 13.13.15 # But with Rockbox we're committing code multiple times per day, from many directions. 13.13.25 # still doable 13.13.47 # I envisage a script that connects to each dev's git repo 13.13.53 # Technically speaking, anything is doable. 13.13.58 # and reads a file containing a list of branches to pull 13.14.21 # all the dev has to do is put a branch in there for it to get pulled to the master repo 13.14.32 # the script checks say every 15 minutes ? 13.14.55 # bit rough and ready, but it would work 13.14.58 # linus actually talks about that in the vid too...he mentions a part about when there are issues with the merge....but, yes, atm in my research, it seems like it puts more load on the maintainer(s) which i dont like...i need to research a bit more to figure out exactly the amount we are talking here, though 13.15.05 # This depends on single development machines, or single internet access points (or domain names), etc. 13.15.21 # why not git-push? 13.15.40 # * scorche adds git-push to the list of things to check out 13.15.50 # * GodEater hasn't come across git-push yet either 13.15.59 # those who already have svn commit access could push to a central repo 13.16.02 # git-push brings us back, mostly, to why not just svn-git? 13.16.07 # Or was it git-svn? 13.16.12 # git-svn 13.16.49 # dionoea: Do you remember what the problem was regarding xobox on archos? 13.16.54 # I don't understand the context of a "ref" with git-push 13.17.07 # If it was speed, could it be that your recent changes make it run fast enough? 13.17.43 Join Arathis [0] (n=doerk@p5484902A.dip0.t-ipconnect.de) 13.17.46 # I assume git-push is the equivilant of git-svn excecpt the mster would be running git instead of svn? 13.17.58 # JdGordon: that's not clear from it's man page 13.18.16 # at least, not to me :) 13.18.38 # * GodEater accepts he may be both stupid and ugly as definted by LinusT 13.18.43 # *defined 13.19.03 # oye...according to him, if you disagree with him, you are ugly and stupid 13.19.17 # except if you work on the KDE project :) 13.19.20 # GodEater, Scorche: Just to clarify, I do see the advantages of git in a lot of situations, I'm just not sure Rockbox is one of them. 13.19.25 # I honestly cant see how having 1 maintainer could possibly work for any project, that would be a full time job and a bloody nighmare 13.19.45 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb) 13.19.50 # JdGordon: i was mentioning that before, but as i said, i dont really knwo how much work this would be 13.19.51 # JdGordon: I'm actually considering emailing him and asking him about rockbox's situation :) 13.20.04 # JdGordon: In software that does actual release versions, someone or a team of someones has to define what is a release, and you have programmers with objectives. In that situation it makes a lot more sense. 13.20.13 # GodEater: same, although it would be easier to just ask him to come in ehre ;) 13.20.18 # scorche: if all the devs went with git-svn, there's be no extra work :) 13.20.30 # but we wouldn't get all the benefits either 13.20.45 # GodEater: it has its disadvantages....as well as performance issues comparatively 13.21.05 # I've not used it yet, so couldn't comment 13.21.13 # I've only used git on it's own 13.21.19 # hmm... seems git can do a push anyway 13.21.36 # although it seems more cmplicated than svn commit 13.21.42 # GodEater: go for it :) 13.22.44 # JdGordon: well, actually, he is saying most things in his talk about a project that makes rockbox look miniscule, but i really need to learn a bit more about this and get my hands dirty in it 13.22.57 # this could be interesting.. autoaccepting git deltas from trusted email adresses.. 13.23.02 # JdGordon: wouldn't be much point given I don't have commit access to rockbox :) 13.23.43 # scorche: he says git is better for everyone... so he shold be able to say the best situation for us 13.24.01 # of course....i just wish there were more details 13.24.05 # * GodEater is actually having difficulty tracking down the big L's email address 13.24.23 # linus.torvalds@linux.org? 13.24.23 # is he still at transmeta ? 13.24.33 # GodEater: he likely has many for each project 13.24.55 # the video had an email address in the cover slide 13.25.00 # actually - perhaps emailing him is the wrong thing to do 13.25.01 # Does git work on Windows? Judging from references to "hard links" when cloning repos I'm guessing not. 13.25.14 # perhaps we should direct an email at the git mailing list (assuming there is one) 13.25.22 # GodEater: No, OSDL IIRC 13.25.25 # safetydan: no idea 13.25.33 Quit linuxstb (Read error: 110 (Connection timed out)) 13.25.33 # doesnt sound to bad an idea if you dont mind it being public 13.25.36 # linuxstb: thought so 13.25.49 # doesn't bother me, does it bother anyone else ? 13.26.26 # Asking about how appropriate git is for our project you mean? 13.26.44 # git sort of works in cygwin, but it's painful 13.26.45 # yes 13.26.45 # GodEater: More than you wanted to know.... http://www.catb.org/~esr/faqs/linus/ 13.27.18 # linuxstb_: far more :) 13.27.22 Quit driver8` ("I said the dingo ate your baby") 13.27.31 # Llorean: that yes was at your question by the way 13.27.32 # GodEater: I'd say go for it, but be sure to mention that we do distribute binaries, the majority of our users need binaries, so there's very definitely one final repository despite distribution. 13.27.42 # GodEater: I figured it was, I was just slow typing a response. 13.28.28 # Llorean: it was the main part of the question I was going to ask :) 13.28.42 Join AceNik [0] (n=AceNik@203.145.159.42) 13.28.44 # GodEater: yes lack of Windows support is a bit of a downer. There are alternatives though like Mercurial that do work on Windows. 13.28.44 # GodEater: this is likely much more than he will want to answer (he is a busy guy), but also ask about how much load is ont he maintainer(s) of the head 13.29.17 # scorche: going with the mailing list approach - so hopefully we'll receive answers from *someone* if not Linus himself. 13.29.23 # how well does it do deltas on binaries? we could set up a branch for the zips instead of serving them from the www server 13.29.24 Quit billytwowilly (Read error: 104 (Connection reset by peer)) 13.29.28 # I thought rockbox itself didn't even compile in windows unless you used cygwin 13.29.38 # http://marc.info/?l=git <-- mailing list archive for anyone interested. 13.29.54 # * amiconn wonders why hardlinks shouldn't work on windows 13.30.01 # he mentioned mercurial in the talk as another distributed system, but doesnt much into it 13.30.04 # Galois: Yes, but many people use TortoiseSVN or something similar, so they don't do anything *except* the compiling in cygwin 13.30.13 # At least on W2K and higher they're definitely possible 13.30.23 Join billytwowilly [0] (n=chris@S01060015e91ce305.cg.shawcable.net) 13.30.34 # amiconn, true, ntfs supports them 13.30.35 # it would be worth finding out if git is able to be gotten for cygwin as well (without compiling) 13.31.04 # Take care everyone. 13.31.05 # scorche: Mozilla project went with mercurial after much discussion partly because of the poor Win32 support of git 13.31.11 # * amiconn uses them (so-called 'junctions') to have only one testfile dir shared between all his sims 13.31.15 # * JdGordon waits for bagder to come in a say "we are not bloody converting again"... 13.31.22 # haha 13.31.43 # well, Zagor started this whole talk! =P 13.32.18 # heh, are you still talking about this? :) 13.32.23 # yes! 13.32.26 # repository checkout with git-svn takes a while on my poor laptop and that's just for the tip. I'd hate to hink how slow a full history conversion would be. 13.32.27 # * Zagor whistles innocently 13.32.32 Quit Kanashimi ("-") 13.33.10 # * GodEater will post his question at pastebin or equivalent before he sends it. 13.33.27 # safetydan: yeah...that would be one of the pitfalls of git-svn....honestly, provided the load isnt too much ont he maintainer(s) of the head, and we handle it right, i dont see a problem with going full git 13.34.06 # if we convert, we'll do it fully. no half-measures. 13.34.21 # scorche: I think it'll slow down the 'commit' (pull?) rate quite a bit. 13.34.35 # would anyone tell m , how to redesign a plun originally made for another player ? 13.34.43 # Unless there's a solid group of maintainers of the live branch around the clock 13.35.07 # AceNik: a what? 13.35.12 # Llorean: I would guess anyone with current access would become a maintainer? I couldnt see how i would work without that 13.35.12 # LinusN: Plugin maybe? 13.35.18 # AceNik: You mean how to make a plugin which works on one Rockbox target work on another target? 13.35.27 Quit spiorf (Read error: 110 (Connection timed out)) 13.35.47 Join spiorf [0] (n=spiorf@host55-222-dynamic.8-87-r.retail.telecomitalia.it) 13.35.51 # JdGordon: That somewhat goes against the philosophy of git, though. 13.36.01 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb) 13.36.02 # amiconn: i've never tried it on archos. 13.36.22 # Llorean: no, i dont think so, the article I read says git has a way to push your commits up 13.36.26 Quit spiorf (Client Quit) 13.36.38 Join spiorf [0] (n=spiorf@host55-222-dynamic.8-87-r.retail.telecomitalia.it) 13.36.49 # JdGordon: It has a way to push, as I said, it's against the philosophy, not necessarily the capability 13.36.50 # so sorry my apologies 13.37.23 # * JdGordon wonders how git blame is supposed to work if there is only 1 person doing real commits 13.37.46 # ok more precise & correct : how o you resize & recode a plugin designed for another player , to work on the desired player 13.38.00 # AceNik: Do you know C? 13.38.17 # JdGordon: those "real commits" are tagged with the developer name, aren't they? 13.38.17 # yup a little 13.38.29 # isnwhat you 13.38.40 # AceNik: You basically need to open up the source for the plugin, and find the player-specific bits (screen positioning, keymaps) and write new ones for your player. 13.38.48 # Zagor: not by the maintainers name? 13.38.50 Quit Arathis ("Bye, bye") 13.38.53 # isn't what you call "real commits" rather more like a "collection" of other peoples' commits? 13.39.10 Join Arathis [0] (n=doerk@p5484902A.dip0.t-ipconnect.de) 13.39.17 # by real commit, I mean the equiv of svn commit 13.39.20 # AceNik: Or make the code more generic, so it works on any player (e.g. position items in terms of a percentage of the LCD's width, rather than an absolute number of pixels) 13.39.46 # JdGordon: well there is no equivalent as I understand it, since there is no One True Repository 13.40.07 # Llorean:i have opened up the .diff file made as a plugin by someone, i uess there are keymaps first , then the code of the game exists , so if i just add a section to have the H10 keys will it work 13.40.14 # Zagor: but, we would have to have a "head" that we use for binaries and such 13.40.19 # on the other hand I could be confused, I haven't studied git much yet 13.40.27 # dionoea: Hmm. I think all plugins should work on all targets whenever that makes sense and is technically possible 13.40.36 # scorche: indeed. we pick a maintainer we trust. just like linux. 13.40.40 # so going by that, doesnt it sound like the repo would never actually be updated by anyone except the mainter? hence git blame would be pointless other than showing when a line was chagd 13.40.54 # AceNik: Is your screen bigger or equal in size to the screen of the player it's meant for? 13.40.56 # AceNik: Don't edit the diff, edit the source code after you've applied the diff. How much you will need to change to make it work on the H10 will depend on how generic the plugin is. 13.41.02 # thats what i thought , that all pluins would work , but there is some work with the resizing 13.41.13 # JdGordon: I don't think "git pull" changes the author tag (or whatever it might be called) 13.41.16 # (Example for #1 not applicable: iriverify only makes sense on irivers, example for #2 not applicable: doom won't run on archos) 13.41.26 # Zagor: I think we'd need a minimum of about 5-8ish maintainers. 13.41.31 # Zagor: ok 13.42.05 # Zagor: Just to be sure things don't slow down when that one maintainer's life gets busy, and so that things can happen during 'night' in their timezone. 13.42.13 # Rockbox seems to be characterized by lots of smallish commits. 13.42.20 # we'd need more maintainers than that... does any of the current commiters not have a job? 13.42.29 # firstly there aint no source right now , it a plugin game called spider solitaire & the author made a .diff nopthing else , it might be making files i guess 13.42.30 # Llorean: yes that is a question mark about git. does it, in essence, make a bottleneck of a single maintainer 13.42.49 # check this out http://www.rockbox.org/tracker/task/6599?histring=spider 13.42.50 # that may be fine for linux, but I don't think we'd want that for rockbox 13.42.50 # i dont see an issue with lots of maintainers on the head (and would probably prefer it), but i still need to figure out exacttly how much load we are talking about 13.43.08 # scorche: load? 13.43.11 Quit DerPapst (Read error: 104 (Connection reset by peer)) 13.43.32 # amiconn: I agree. (I didn't code the plugin in the first place) 13.43.51 # according to http://git.or.cz/course/svn.html#share git can automatically accept incoming email changes, so if we used that we would only ned to restict the emails to known addresses, and as long as the change actually merged ok we wouldnt need maintainers 13.43.54 # as in how much work that(those) maintainer(s) would have to undertake....what they actually have to do specifically 13.43.57 # If one of my commits yesterday broke anything i'd be glad to fix it though 13.44.02 Quit GodEater ("CGI:IRC") 13.44.15 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-24c3a53c2c55eb59) 13.44.22 # JdGordon: yeah, that sounds just about like our current system 13.44.29 # http://pastebin.ca/536426 <-- email I plan to send. Comments / suggestions please. 13.44.58 # but then we're back to a One True Repo again, which I thought was antithetical to the git philosophy? 13.45.23 # I should probably read more about it before discussing... 13.45.41 # Zagor: we will need a head repo regardless for binaries, as i said...this is more about how the developer is affected 13.45.49 # yeah 13.46.06 # bbl 13.47.03 # people who read the logs sure will have an eyeful today... =) 13.47.10 # indeed :) 13.47.26 # and those who dont, will likely be lost 13.47.36 # anyone like / dislike my email ? Do I hit send with it as is ? 13.48.06 # GodEater: I think it's mostly good, based on my limited knowledge. 13.48.23 # sent then 13.48.36 # GodEater: add to 1) that it is also for support purposes 13.48.40 # oh...bah 13.49.17 # scorche: I'll bear it in mind in case anyone asks ? 13.49.19 # :) 13.49.34 # can i have a link to the thread when i appears, so i can watch it? 13.49.46 # pondlife: (2 huors later :p ) all this git talk got me to fix that problem with 4 lins of code :D new patch up i a min 13.50.25 # certainly 13.50.34 # bah... why did I ever post that git talk link in here :/ 13.50.51 # http://marc.info/?l=git&m=118095771216110&w=2 13.50.51 # cries 13.51.01 # petur: I did it first :) 13.51.06 Quit safetydan (Remote closed the connection) 13.51.09 # i cant get any player to work on my photo ipod that allows for fast forward and rewinding ! :( 13.51.10 # I think =/ 13.51.19 # GodEater: when? 13.51.25 # can someone help me with this sokban patch 13.51.53 # ah...so it was petur who started it =) 13.52.00 # GodEater: i dont reall tihnk the binaries need is the thing stopping git for us 13.52.03 # petur: 08:58 13.52.09 # today? 13.52.12 # yep 13.52.14 # spider solitaire? 13.52.14 # * scorche apologizes for pointing the finger at Zagor 13.52.23 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb) 13.52.30 # GodEater: http://www.rockbox.org/irc/reader.pl?date=20070603#21:02:41 13.52.46 # anyone know a movie player for rockbox that allows for fast forwarding and rewinding of a movie otherwise the players are just pointless to watch full length films... 13.52.49 # petur: ah - my mistake - but I'd have posted it anyway :) 13.52.50 # petur: his mention was before you, but his was largely ignored 13.53.23 # * scorche shuts up 13.53.23 # NIXON-Trent: Watching a film *requires* fast forward and rewind? 13.53.27 # I wish I never started it 13.53.30 # Startling, I've never had to in the theatre... 13.53.37 # * Llorean learns something new every day. 13.53.45 Part AceNik 13.53.46 # lol 13.53.48 # Llorean: hahahaha 13.53.51 # gotta love smarty pants :) 13.54.02 # NIXON-Trent: Fast Forward and Rewind will exist some day, mpegplayer isn't "finished" in terms of features yet. 13.54.13 # well im not watching it on a large screen where i've dedicated 2hrs of my time to watch the movie 13.54.25 # petur: well, if it leads to a better "life", you can be the hero =) 13.54.28 # You're perfectly welcome to work o n'em. 13.54.34 # on 'em. 13.54.37 # its more hey im traveling on a train or car and can watch part of the simposons eposide.. 13.54.50 # i would if i knew the language u guys code in. 13.55.22 # i spent 2hrs trying to get podzillas player working but its a peice of crap and i am still un able to load or use it to play any file :/ 13.55.26 # NIXON-Trent: well, the simple fact is that coming here and complaining will not do anything about it...either learn C and program it yourself, or wait patiently 13.55.36 # did i complain 13.55.41 # no i asked for an alternative... 13.56.07 # and i was replyed to with a uneeded reply. 13.56.22 # 21:53:16 ] NIXON-Trent: Watching a film *requires* fast forward and rewind? 13.56.31 # NIXON-Trent: You could our mpegplayer "pointless" 13.56.43 # I provided my opinion that fast forward and rewind were not, in fact, "necessary" 13.56.48 # refering to my situation. 13.57.13 # but as we continue this converstation i guess theres no help here to be found :/ 13.57.23 # if you consider it "pointless", then dont use it and get a device that is designed for watching videos 13.57.24 # If someone implemented fast forward or rewind, we'd included it 13.57.40 # ipodlinux's mv player can.. 13.57.42 # :/ 13.57.47 # then use that 13.57.48 # its just i've found it a pain to get working 13.57.55 Quit linuxstb (Read error: 113 (No route to host)) 13.57.57 # NIXON-Trent: And have you noticed any OTHER differences in their video player? 13.58.18 # it works 13.58.23 # then use it! 13.58.28 # and it stops, plays, fastworward and rewinds 13.58.34 # then use it! 13.58.41 # And it uses a different file format, so their code is useless to us. 13.58.48 # i'm trying to but having problems getting it to work on my ipod.. 13.58.58 # well, go to them about that then 13.58.59 # first reply is in, it's not that helpful though =/ 13.59.02 # so because i like rockbox better decided to ask if there was an alternative. 13.59.05 Join linuxstb__ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net) 13.59.11 # GodEater: Is there a threaded view of i t? 13.59.18 Quit ljcrabs2 ("Leaving") 13.59.20 # sorche keep out of it :) 13.59.30 # NIXON-Trent: Again, that sounds like you're assuming that if there was a better video player, for some reason we'd intentionally not include it? 13.59.45 # u dont need to be llorean's instant repeat. 13.59.51 # i am not... 14.00.01 # Llorean: the link I posted above is all there is 14.00.19 # GodEater: MIssed that, thanks, found it. 14.00.28 # i dont know but if ipodlinux can make seeking functions im suprised rockbox hadnt :) 14.00.30 # thats all 14.00.37 # NIXON-Trent: They use a different video format 14.01.13 # duh 14.01.20 # So you can't compare them on the same grounds 14.01.21 # NIXON-Trent: we are a project which consists of volunteers only...not all of us have the time to implement it 14.01.24 # Seeking works differently. 14.01.44 # Our video format is more compressed and offers a lot of benefits, but seeking code takes different work. 14.01.46 # anyone know how much disk space havig the entire hitory locally is going to take? 14.01.47 # scorche i do know that :) 14.02.05 # * JdGordon doesnt like the idea of gb's of data instead of the 60mb of source code 14.02.15 # well, 30 and the .svn dupe 14.02.56 # JdGordon: well, he mentioned it compresses it and that each branch is a tiny little file, but other than that, i dont know 14.03.39 # anyone with a fast enough computer wanna pay guienea pig and import all our svn? 14.03.55 # classify fast. 14.04.13 # UBER-freeking-FAST! 14.04.14 # i will tomorrow, likely...just to play...i have been putting off sleeping ever since this started though 14.04.46 # If i had linux on my brothers comp id do it there... little buggers comp shits on mine :p 14.06.48 Join AceNik [0] (n=AceNik@203.145.159.42) 14.06.56 # vmware? 14.07.19 # hey guys what does "skipping trailing CRs " mean 14.07.23 # coliux, its faster :D 14.07.32 # * scorche shrugs 14.08.38 Quit linuxstb_ (Read error: 113 (No route to host)) 14.08.46 # AceNik: http://en.wikipedia.org/wiki/Newline 14.08.56 # jhMikeS: you round? 14.10.26 # dionoea: Do you think xobox might work in pure b&w? Or does it need greyscale at least? 14.10.42 Join barrywardell [0] (n=barrywar@host-194-46-234-199.dsl-ie.utvinternet.net) 14.10.47 # pure b&w should work 14.10.51 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb) 14.11.03 # You basically would have black dots moving on a white background 14.11.08 # black borders 14.11.26 # and draw the player's cursor using some inverse color or something like that 14.11.41 # and if it comes to it, you can do something like bubbles did with different designs per group 14.14.05 Quit printfXh4 ("Leaving") 14.16.11 # * scorche finally gets a chance to go to sleep and seizes it 14.16.57 # JdGordon: Sorry, back now. Still good to try the new patch? 14.17.16 Quit linuxstb__ (Read error: 113 (No route to host)) 14.17.51 # please :) 14.18.52 # OK, building... 14.19.20 # * JdGordon says bugger it... im importing svn into git 14.19.30 # ^ is my rough timestamp for the start 14.19.38 # JdGordon: Nope, still broke :( 14.19.47 # 33/13500 commits done 14.19.54 # pondlife: how? 14.20.16 # same problem? 14.20.25 # Problem #1: Scroll down a folder - the cursor bar can't get onto the bottom 2 lines. 14.20.51 Join kaaloo [0] (n=luis@rue92-3-82-232-48-241.fbx.proxad.net) 14.21.13 Part kaaloo 14.21.14 # Ah, is this intentional? It's just that it maintains 3 lines lookahead at the top and bottom. 14.21.37 # yeah, we try to keep the cursor off the top and bottom 14.21.44 # Is that new behaviour? 14.21.48 # no 14.22.15 # the remote lcd seems to have problems in the main menu though :( missed that testing 14.23.03 # 253 commits done, 2mb disk usage 14.23.23 # How does it decide how much screen to "reserve"? It's quite a lot....about a third.. 14.23.42 # thats exactly what it should be 14.23.54 # no... 14.24.01 # min of 2 lines, up to 1/4 of the screen 14.24.28 # I get 3 lines when moving up, or 2 when moving down. 14.24.40 # Would prefer a max of 2.. 14.24.45 # Or not at all. 14.24.52 # But that's just me. 14.25.10 # its better with a bit of a gap, but yeah, would be better if both were 2 or 3 14.25.53 # Ah, I've found a bug which *is* new though. 14.26.12 # :( 14.26.55 # Navigate into a dir with only 3 entries, then into a subdir of that with lots of entries. Move down lota (so you're near the end) and press LEFT to go back up. 14.27.17 # The cursor bar end up off the top of the screen ! 14.27.49 # That dir backtrachas been broken for months. It used to work properly in the past :/ 14.27.52 # yeah, fixed that 14.27.53 # You need to make it scroll when in the subdir. 14.28.20 # This seems to be a new bug though. Wasn't in the 7th March build at least. 14.28.22 # Backing out of a subdir should place the cursor on that very subdir 14.28.35 # Indeed. it does, but doesn't ensure it's visible. 14.28.38 # For quite a while it doesn't do so anymore 14.29.01 # thats fixed 14.29.09 # New patch? 14.29.40 Quit lids (Remote closed the connection) 14.30.13 Quit ompaul (Connection timed out) 14.30.13 # * JdGordon wonders why there is no files outside of .git yet after 500 commits... 14.30.16 Quit intgr (Read error: 104 (Connection reset by peer)) 14.30.19 Join intgr [0] (n=ack@blip.juffo.org) 14.30.29 # unless the cvs->svn import did a crazy amount of empty commits? 14.30.54 # hmm. I'm not sure either reply I've had from the git list is much help so far. 14.31.00 # JdGordon: Pop it on http://www.rockbox.org/tracker/task/7258 14.31.30 # :( 14.32.20 # done 14.33.04 # Building... 14.33.16 # JdGordon: are you trying out git-svn ? 14.33.44 Quit AceNik ("bye guys have fun , enjoy !!!!!") 14.33.48 # GodEater: no, im importing the current tree into git to see what sort of size we are looking at 14.33.53 # ah 14.34.04 # it wasn't that huge when I did it 14.34.11 # its pretty fast :) 14.34.42 # 10% done 14.34.49 # yeah, I thought so too 14.35.20 # so shuold be done mby midnight-ish :p 14.35.59 # * GodEater has no clue what the time is where JdGordon is 14.36.05 # 22:36 14.36.06 # JdGordon: Much better 14.36.32 # you must be doing something other than a "git add ." 14.36.38 # it took less than a minute for me :) 14.36.46 # GodEater: git-svnimport 14.37.07 # ah 14.37.08 # * JdGordon appologises to haxx for the download wastage 14.37.28 # can't try that here - no proper intermaweb access 14.37.30 # JdGordon: LCD seems ok too, is that menu fixed? 14.37.42 # not quite 14.37.43 # 30MB repo when done with git add though 14.37.54 # which is expected... 14.37.54 # No probs, lunch time for me 14.41.00 # * Llorean is tempted to make use of the word "allow" a bannable offense in the forums. 14.41.24 # Well, not really, but reading posts that say "Why doesn't Rockbox allow me to do Blah" gets irritating after a while 14.41.32 # :) 14.41.42 # JdGordon: can you use git-svnimport on an already checked out local repository ? Or does it have to come from the svn master ? 14.41.43 Join Rob222241 [0] (n=Miranda@p54B1497E.dip.t-dialin.net) 14.41.44 # Like it's a matter of that feature being out there, waiting, but Rockbox is standing in the way saying "No, it's there but you can't use it" 14.42.04 # GodEater: from the master i assume, it pulls the whole history also 14.42.23 # ah 14.42.32 # doing the test because apparently git stores that everywhere... 14.42.45 # so want t see how much disk space we woiuld all waste 14.42.57 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul) 14.44.05 # Well, for those of us using the vmware image, it's a non-issue, I think. :-P 14.44.14 # The image already reserves more disk space than we're likely to need for it. 14.44.32 # its a non issue if you need 500mb extra disk space? 14.44.38 Join XbooX [0] (n=xboox04@206.186.45.240) 14.44.54 # The virtual disk file is like... 2gb already, by default 14.45.11 # If I need to use 500mb more of that already-used 2gb, it's not a major concern. 14.45.27 # why is it so big? 14.45.33 # Couldn't say. 14.45.34 # ask scorche :) 14.45.44 # thats crazy :p 14.46.54 # Llorean: found a better view of the git list : http://news.gmane.org/gmane.comp.version-control.git 14.47.13 # GodEater: I thought you mentioned a second response 14.47.30 # though for some reason it's still not showing the 2nd reply I had, and he definitely copied the list on it... 14.47.35 # Aaaah 14.47.43 *** Saving seen data "./dancer.seen" 14.48.09 # I'll pastebin that too :) 14.48.26 # The 2GB virtual disk is actually rather tight 14.48.37 # amiconn: Really? 14.48.44 # I hadn't checked, but I'd assumed there was some elbow room. 14.48.49 # http://pastebin.ca/536526 14.48.52 Join lee-qid [0] (n=liqid@p549677a0.dip.t-dialin.net) 14.48.54 # Try building 10+ sims, enable ccache etc.... 14.49.13 Quit GodEater ("CGI:IRC") 14.49.14 # I'd rather reserve 5..10GB for a full rockbox dev environment 14.49.26 Join GodEater [0] (i=c2cbc95c@gateway/web/cgi-irc/ircatwork.com/x-e89c2ac694d8c41f) 14.49.28 # As Zagor said, disk space is cheap... 14.49.33 # * GodEater would just rather use linux 14.50.26 # * Llorean still keeps delaying installing Linux on his laptop. 14.50.29 # The OS doesn't matter here. The VMware image _is_ linux 14.50.43 # I wouldn't mind reserving 5 or even 10gb for it either. 14.52.05 # but you're having to "reserve" space with vmware 14.52.17 # if the entire machine is running the OS, you don't have to reserve it 14.52.22 # True 14.52.22 # you can use the entire disk if you need to 14.52.25 # or not if you don't 14.52.41 # My real problem is hardware I haven't tracked down a way to use under linux yet. 14.52.58 # laptops are a PITA like that =/ 14.53.09 # you could use linux on a second machine ... 14.53.11 # Llorean: really new or really old hardwrae? 14.53.42 # JdGordon: HTC Faraday / 2125 used over USB for internet connection while travelling. 14.53.57 # So, fairly new, but it's also windows mobile which makes things more exciting. 14.53.59 # * GodEater pretends to know what that is 14.54.08 # GodEater: 's a phone 14.54.11 # ah 14.54.14 # my verizon phone over usb works fine in linux 14.54.17 # EVDO 14.54.51 # I imagine mine might, I just haven't gotten around to putting together a liveCD of some sort and testing to be sure. 14.55.08 # But I haven't found any specific accounts of users with it yet either. 14.55.20 # * petur made the switch to linux on his new laptop 14.55.29 # * GodEater needs to update his PS3 linux install to get the wifi working 14.55.42 # GodEater: The default for workstation vmware disk images is to grow. For some reason our dev image uses the reserve-all type... 14.56.08 # amiconn: isn't that because it was created in VMWare server ? 14.56.15 # no 14.56.30 # Even the server allows for growing images, it's just not the default there 14.56.46 # 23% done.... 14.56.50 # Only ESX server doesn't, but those images aren't compatible with workstation or player anyway 14.56.53 # amiconn: Were you working on building your own vmware image? 14.56.58 # yes 14.57.10 # I still have 2 problems with it 14.57.33 # (1) Even though it doesn't have X11, it's bigger than the one we're currently distributing 14.58.00 # start with a ultra-minimal slackware setup :) 14.58.01 # (2) For some reason building the html manual fails, although the same thing works on my full fledged debian vm 14.58.29 # But it does build all targets and pdf manuals, and it does build win32 sims 14.58.42 # ESX Server does allow for image growing, it's just not compatible with Server / Workstation because it requires a SAN 14.59.05 Quit Rob2222 (Read error: 110 (Connection timed out)) 14.59.09 Join AceNik [0] (n=AceNik@203.145.159.42) 14.59.17 # does anyone want to test the fix for the wierd sansa accel issue before i commit? (someone with an ipod would be excelant..) 15.00.13 # Llorean: phones just work like old dial-up modems. you say "ATDT*99#" to it and you get a ppp response back. 15.00.28 # so I'd be surprised if the HTC doesn't work 15.00.28 # good ol' Hayes command set 15.00.29 # Zagor: Ah, interesting. 15.00.38 # those were the days :) 15.00.45 # Zagor: I had to jump through about 50 driver hoops to get it working in windows, it'd be lovely if it's simpler under linux. 15.00.50 # I bet it is 15.00.53 # my phone doesn't work in windows 15.01.09 # I mean, I think it might work if I jumped through 50 driver hoops, but fedora has the drivers built in 15.01.17 # Llorean: first test is to plug it into a linux box and see what it is identified as 15.01.38 # you hook up the phone and /dev/ttyACM0 pops up, then you pretend it's /dev/modem like Zagor said 15.01.55 # the only difference is it's faster than dialup 15.08.01 # can someone with an ipod plase check http://www.rockbox.org/tracker/task/7242?getfile=14051 and tell me if it causes/fixes any scan accell issues? 15.11.50 # pretty please.... 15.16.56 # JdGordon: what am I supposed to be looking for after applying the patch? 15.17.12 Join darkless [0] (n=darkless@62.79.44.48.adsl.vby.tiscali.dk) 15.17.25 # anything odd in the lists when scrolling 15.17.33 # wierd acceleratino 15.19.23 # building... 15.20.20 # amiconn: from what I can see, you can specify that disks grow in vmware server, but still only up to a given size. So you might as well pre-allocate them as it's faster for the VM when it's running. 15.20.43 Join lids [0] (i=lds@gateway/tor/x-d4a50084e64c3cbc) 15.21.26 # 3rd reply on the git list. Perhaps the most helpful so far. 15.22.03 # suggestion is to look here for how to set up a "central" git repository : http://repo.or.cz/ 15.24.44 Part XbooX 15.25.43 # GodEater: I doubt that disk growing has a noticeable impact on performance in vmware. (Since it doesn't grow very often) 15.26.07 # JdGordon: it seems to behave the same as before 15.26.16 # ok great 15.26.20 # will commit soon then 15.26.53 # * petur wonders if "same as before" means with the same bugs :p 15.27.32 # that's kind of what I was asking, because I wasn't aware of any bugs with the old behavior 15.28.47 # JdGordon: I can't test now (but will later), but does it still accelerate on the ipod when moving the wheel slowly? 15.28.57 # thats what im asking about 15.29.04 # it shuoldnt... 15.29.51 # Galois: What happens when you scroll slowly - e.g. one line every second or two seconds. 15.30.03 # with the patch, scrolling once around the wheel advances exactly 18 lines, whether it's fast scrolling or slow (ipod nano) 15.31.15 # here we go, let's try changing the scroll speed in settings 15.32.41 # oops, wrong meaning of scroll 15.34.16 # this is not exactly scientific, but if I scroll super slowly, one line per two seconds like you suggest, it does scroll more lines 15.34.22 # 21 lines per revolution instead of 18 15.35.44 # damn, ok 15.35.53 Join webguest08 [0] (i=3bb72d9a@gateway/web/cgi-irc/labb.contactor.se/x-93c919679a8ac9fb) 15.36.08 # hello 15.36.18 # I have a problem 15.36.40 # just ask the question 15.36.40 Quit webguest08 (Client Quit) 15.36.56 # also fine... 15.37.05 Join saratoga [0] (i=9803c50e@gateway/web/cgi-irc/labb.contactor.se/x-15a7cba77a02e29e) 15.37.17 # Galois: are you leaving your fingure on the wheel when you do this? 15.37.22 # yes 15.37.36 # JdGordon: I just tried it without the patch, and it's *much* worse without the patch than with the patch 15.37.37 # can you tes releasing? 15.37.45 # ok great 15.38.07 # without the patch, it skips lines when I scroll slowly 15.38.18 # linuxstb: that sound alright to commit? 15.38.32 # with the patch, it doesn't skip any lines, but it still changes lines at a slightly faster rate 15.38.52 # as long as its no skipping when it shhuoldnt 15.39.03 # what about going round quickly? does it skip as expected? 15.39.14 # going round quickly is the same either with or without the patch 15.39.23 # k great 15.39.27 # the skipping lines stuff only happens without the patch and if I don't release my finger 15.39.27 # * JdGordon commits 15.40.24 # JdGordon: I'll let you know later when I test it ;) 15.41.25 # JdGordon: But could you explain the algorithm in the commit message? 15.41.52 # the action code now checks the time difference 15.41.52 # between events to decide if its a repeat or not 15.41.57 # good enough? 15.42.45 # So it ignores the "repeat" flag? 15.43.02 # it never checked it... it always went by the action value, not he button value 15.43.33 # im not 100% sure this doesnt break other code though... 15.43.35 # * JdGordon checks 15.44.14 # it wont :) 15.44.36 # I'm not sure I understand - is the action code (i.e. the code to turn button events into actions) checking the time difference, or is the list code? 15.44.41 # and from action.h... "#define ACTION_REPEAT 0x2 /* action was repeated (NOT button) */" 15.45.44 # linuxstb: both, the list wont unless the action was repeated, the list then checks to see if its accelerating yet, and if it is, if its time to increase the skip 15.47.07 Join Bagder [0] (n=daniel@rockbox/developer/bagder) 15.47.34 # OK, so back to my first question - is the time difference between button events the only way to trigger a repeat action, or is the repeat flag from the button driver also used? 15.47.37 # one wifi router dies, another one starts 15.48.19 # linuxstb: only the time diffrence (assumign the 2 actions where the same value) 15.48.41 # pressing stop and left (fas enough) would cause a repeat 15.48.45 Quit AceNik ("bye guys have fun , enjoy !!!!!") 15.48.51 # would appear as a repeat 15.49.35 # OK, thanks And my last question is what is the time difference that triggers an action repeat? 15.49.39 Nick idnar_ is now known as idnar (n=mithrand@unaffiliated/idnar) 15.50.46 # HZ/10 15.51.00 # that might need to be tweaked... 15.51.07 # I think that will work better on ipods (ignoring repeat events), and it's easy to accidentally untouch the wheel when scrolling. 15.51.22 # ^because it's easy 15.51.48 # ok, so commiting... 15.52.02 Join AceNik_ [0] (n=AceNik@203.145.159.42) 15.55.15 Part AceNik_ 15.55.30 Join AceNik_ [0] (n=AceNik@203.145.159.42) 15.55.30 Quit saratoga ("CGI:IRC (EOF)") 15.55.32 # * amiconn wonders why we need an extra repeat mechanism 15.55.54 # Wouldn't it be sufficient to change the way the ipod wheel fires repeat events? 15.58.54 Quit maffe (Remote closed the connection) 15.59.00 Quit ompaul ("reboot, rebuild, reinstall, reconfigure") 16.00.38 # amiconn: without it in the action code, the acceleration in the lists isnt possible, unless we check it there also, so then there is no point moving it 16.01.04 Part AceNik_ 16.01.13 # ... and it was the sansa wheel which brought this bug up.. not the ipod 16.01.47 # 52% completed... looks like im sleeping with the comp on tonight :'( 16.03.09 # * Nico_P had a long read in today's logs 16.03.14 # interesting one too... 16.03.25 Join Rincewind [0] (i=IaK3k8RI@nat-wh-1.rz.uni-karlsruhe.de) 16.04.02 # Bagder: thoughts about git ? 16.04.13 # not really 16.04.22 Quit Entasis (Read error: 104 (Connection reset by peer)) 16.04.23 # other than it uses libcurl so of course it is cool ;-) 16.04.40 # wrong question.... thoughts about possibly moving from svn->git 16.04.43 # :) 16.04.44 # ? 16.04.48 # same answer 16.04.50 # JdGordon: How so? 16.04.58 # how so which? 16.05.00 # I don't have thoughts 16.05.01 # Bagder: how about LinusT bashing SVN ? 16.05.06 # we're not linux 16.05.11 # we don't develop like they do 16.05.19 # linus bashed it long before it even existed 16.05.23 # and continues to do so 16.05.28 # no surprise there to anyone 16.06.02 # not saying git is bad, just saying that linus t's opinion on svn is nothing to care about imho 16.06.29 # JdGordon: The repeat stuff... I take it that the button repeat is enough for list acceleration on button-only targets? 16.07.02 # sure... I just find funny how he likes to bash some projects 16.07.12 # he likes to be a prick 16.07.23 # amiconn: no, the list doesnt deal with buttons at all, only the ACTION_STD_NEXT, PREV, so either it checks the time diference there, or relies on something else to do it.. i.e the action code 16.07.37 # I find his attitude and approach annoying in other projects than linux itself 16.08.22 # I can understand that 16.08.25 Quit Zagor ("Client exiting") 16.08.36 # maybe one day he'll bash rockbox :) 16.08.47 # I'm sure he would if he ever used 16.08.49 # it 16.08.56 # he can be added to the official rockbox hitlist :D 16.09.15 # just look at his language and style when he attacked the gnome project 16.09.27 # totally out of order 16.09.49 # something along the lines of "interface nazis", right ? 16.09.55 # what? giving them patches? (alright, he knew they wouldnt be accepted.. but stll)... 16.09.56 # yes 16.10.10 # using lots of foul language, cursing, calling people names etc 16.10.22 # but sure, he did provide a patch 16.10.54 # his approach may be off, but it doesn't make him wrong =/ 16.11.24 # perhaps, but I would've just added him to my ignore list if someone came doing that on one of the projects I cared about 16.11.33 # I'd probably have done the same 16.12.00 # I share some of views of gnome - I don't like that so much of it's interface is locked down and very difficult to change. 16.12.12 # conversely however, I don't like KDE - it goes too far the other way 16.12.30 # but I certainly wouldn't have taken the tack he did with talking to the devs about it 16.12.37 # he is a little arrogant 16.12.54 # was he baited into that argument? or was that just him being a dick? 16.13.18 # mainly him being a dick 16.13.36 Join arkascha [0] (n=arkascha@mailout.imageware.de) 16.13.39 # Godeater: there is gconf-editor on gnome that can change almost anything, it is just a little bit hidden 16.13.57 Part arkascha ("Konversation terminated!") 16.13.59 # pondlife: back from lunch yet? 16.14.00 # Rincewind: that's sort of my point. It's not very well documented. 16.14.11 # * Nico_P is a happy KDE user 16.14.29 # true, I only found it a few weeks ago (using linux for half a year now) 16.14.31 # * JdGordon doesnt think anything linux/foss is documented properly 16.14.32 # I'm aware of it of course - I just don't think you should have to use some odd esoteric config editor to change some apps method of data display 16.14.34 # but I agree he was pointlessly hard on gnome 16.15.26 # JdGordon: curl is! ;-) 16.15.40 # JdGordon: Rockbox is FOSS, and it's pretty well documented these days. 16.16.32 # present company excepted then :) 16.17.12 # and i disagree Llorean, but I cant write user manuals for shit (let alone cmment my code), so ill be quiet 16.17.24 Quit Rob222241 (Read error: 110 (Connection timed out)) 16.17.37 # Well, I use "well documented" in a relative context, in comparison with both FOSS and commercial products. 16.17.48 # ok 16.18.02 # I'm pretty sure it's impossible to write a user manual that actually qualifies as "good" once the complexity of what you're describing gets past the intuitive point, and at the intuitive point you don't need one. ;) 16.18.28 # :) 16.18.53 # * JdGordon fairly sure at least one of his past lecturers will disagree.. but he isnt here 16.20.20 # * JdGordon itching to commit this list patch, FS#7258 needs testing :) 16.20.55 # JdGordon: Yes 16.21.10 # Hang on, will test again... 16.21.16 # cheers 16.21.53 # cant actually remember if anything changed since you left though :p 16.21.55 # Hmm, that's the same patch isn't it. 16.22.19 # Looks like it... was working ok for me with a basic test... 16.22.21 # JdGordon: what do you want tested ? ie what should I look for ? 16.22.30 # Nico_P: any wierdness in the list 16.22.35 # nice n vague :) 16.22.37 # ok I'll test 16.22.47 # ta 16.23.17 # Yep, that third one is the one I tested 16.24.26 # JdGordon: how's the SVN -> git going ? 16.24.26 Join netmasta10bt [0] (n=torment@pool-72-64-226-231.tampfl.fios.verizon.net) 16.24.31 # so the only thing that doesnt work is the first item in multi-line lists is shown on the last line when you press up? 16.24.49 # Nico_P: almost 60% 16.24.51 # after 2 huors 16.25.00 # how much space ? 16.25.08 # 38mb 16.25.17 # wow that's small 16.25.22 # yeah 16.25.28 Quit barrywardell () 16.25.45 # I have no idea how much data im pulling from the server which is a bit annoying, free time only just started :p 16.26.15 # I'm starting to be really interested by git-svn for my GSoC work 16.26.32 # svn merging is scaring me a bit 16.26.59 # you still have the same headache of fixing conflicts... 16.27.19 # * Nico_P just got a super-weird fading pink screen on his gigabeat 16.27.39 # PSOD ? 16.27.53 # haha not far from that 16.28.25 # JdGordon: I can go below the last item in the main menu... I then get "undefined instruction" 16.28.37 # hod you do that? 16.28.39 # howd* 16.28.54 # not hard... go to the last item and press down :) 16.29.07 # or maybe it needs to be with a repeat action 16.29.59 # ! hadnt seen hat before 16.30.05 # nope, no need for that. I go to the last item. press down once: nothing is selected. press down once more: undefined instr 16.30.23 # good thing I tested then :) 16.30.33 # only on lists shorter than the lcd 16.31.25 # yes. the rest seems fine to me 16.32.09 # Nico_P: sometimes when rockbox on the Gigabeat crashes the screen is no longer being updated and then you get a very weird fading screen 16.32.30 Join barrywardell [0] (n=barrywar@host-194-46-234-199.dsl-ie.utvinternet.net) 16.32.34 # win 36 16.32.37 # oops 16.33.11 # markun: it's the very first time I've had that. and there's been a commit affecting the LCD code very recently... *maybe* it's related ? :) 16.35.35 # no, I don't think so 16.35.53 # I had it many times in the early days of the port 16.35.57 Join aliask [0] (n=chatzill@c210-49-190-113.eburwd8.vic.optusnet.com.au) 16.36.16 # ok 16.36.37 # it was a scary experience 16.36.40 # * Nico_P is still shaking 16.37.10 Quit barrywardell (Client Quit) 16.37.47 Join barrywardell [0] (n=barrywar@host-194-46-234-199.dsl-ie.utvinternet.net) 16.40.43 # JdGordon: I think I just saw the same issue on the H300 sim... where you can scroll off the bottom, right? 16.41.19 # yeah, apparnetly I never got it because the root menu exactly fits the screen with the test font 16.41.41 # I'm usingthe Settings menu (it's a bit shorter) 16.41.53 # Weirdly the LCD appears to wrap ok 16.41.59 Join Guile` [0] (n=Guile@88-136-15-165.adslgp.cegetel.net) 16.42.00 # i.e. the remote 16.42.34 # all fixed 16.42.54 # New patch, or are you itching to commit anyway? ;) 16.43.00 Join mpeccorini [0] (n=mpeccori@mail1.theargusgroup.us) 16.43.48 # patch online 16.44.16 # not commiting untill im sure everything i fixed, I and everyone is sick of commits to list.c :p 16.45.06 # Good ! 16.46.35 Join toffe82 [0] (n=chatzill@h-74-0-180-178.snvacaid.covad.net) 16.47.47 *** Saving seen data "./dancer.seen" 16.47.47 # Well that seems ok, but then I never spotted the last problem!! 16.48.13 # How do I display a multi-line item? 16.48.28 # Disk info? 16.48.30 # id3 viewer 16.48.34 # or partition info 16.50.13 # Looks ok 16.51.02 # it doesnt work on the remote wen there is only 3 lines, but that doesnt matter i tinhk 16.54.18 # so, good to go? 16.54.20 # Nico_P: ? 16.54.31 # yes ? 16.55.23 # foun any other problems? 16.55.40 # no. haven't tested the corrected patch yet 16.56.11 # I'm trying to get a VB .NET app I coded some time ago to run with mono 16.56.52 # ok, no prob 16.57.46 # * GodEater had no idea Nico_P was such a masochist 16.58.20 # which part do you find masochistic ? coding in VB .NET or trying t get the app to run on mono ? :) 16.58.39 # the first part is mildy masochistic. The second part is just off the chart :) 16.59.01 # 1 years worth of commits to go 16.59.07 # we'llsee :) 16.59.16 # have you ever tried it ? 16.59.51 # vb.net yes. Trying to get it to run under mono? No. 17.00.14 # JdGordon: the number of commits has been growing much higher in recent months comapred to the first years 17.00.18 # it's an useful program and I don't have windows anymore so I got curious 17.01.12 # petur: yeah, sounds about right, 10k commits up to about this time last year... 5k since then 17.01.15 Join jgarvey [0] (n=jgarvey@cpe-066-057-231-236.nc.res.rr.com) 17.05.02 Quit barrywardell () 17.05.14 # can anyone help me with a really simple script to print the time when a program isnt runnig anymore? 17.05.42 # program-you-want-run && date ? 17.06.07 # program-you-want-run ; date even 17.06.09 # or, if you want the time even if the program dies, then "program-you-want-to-run;date" 17.06.14 # hehe 17.06.16 # * linuxstb was too slow 17.06.17 # haha 17.06.35 # i mean, print the time when a diffrent program finshes.... 17.06.47 # not following 17.06.48 # one thats already bee started 17.06.54 # different-program-your-want-to-run ; date 17.06.58 # hahahaha 17.07.03 # :'( 17.07.06 Part LinusN 17.07.17 # you mean the one you want to check on has already been started JdGordon ? 17.07.23 # bingo 17.07.28 # If the program is modifying files, you can just check the timestamps of those files. 17.07.45 # so you're going to need to check the output of ps, running periodically 17.07.50 # strace -p [pid] ; date 17.07.52 # for the absence of said program 17.07.56 # but it'll take cpu time... :-) 17.07.57 # yeah, that'd do 17.07.57 # could do that, but too hard, dunno what files its going to leave behind 17.08.09 # bagder's is simplest 17.08.19 # Is it too late to stop the program and start it again? 17.08.28 # possibly you should pass some more arguments to strace so it doesn't pick everthing 17.08.45 # and I''d redirect the output of the strace to /dev/null too 17.08.50 # linuxstb: I suspect the program is the SVN to git conversion so probably 17.09.01 # am I right ? 17.09.04 # Nico_P got it in one! 17.09.05 # good guess I think Nico_P 17.10.36 # But yes, the suggestion of "strace -p [pid] &> /dev/null ; date" will work... 17.11.00 # you could probably get away with typing date in the terminal from which you launched the git-svnimport 17.11.11 # will that eat lots of cpu? 17.11.16 # because, you know, terminals are smart and have buffers 17.11.32 # good idea :) 17.12.10 # how about "strace -enone -p [pid] >& /dev/null ;date " 17.12.55 # Galois: if he has a decent bash prompt, he wouldn't even need to type date :) 17.13.28 # * GodEater has a kick ass bash prompt 17.13.33 # so you're suggesting ^Z; export PS1=something decent; fg 17.13.48 # er, not quite no :) 17.13.52 # GodEater: yeah, I thought about fixing that, but does it take affect without reloading? 17.14.13 # well you could do as Galois says - but the ^Z would stop the process running while you make the change 17.14.25 # I'm not sure how important the total running time is to you 17.14.36 # not that important... 17.14.42 # whats 30s when its gone 3 hohuors alreadt 17.14.58 # I could email you my bash prompt script ? 17.15.06 # anyway,. got the strce going, and the buffered date... (some hoping it doesnt ask for input :D) 17.15.16 # na im right 17.15.48 # strace doesn't ask for input, it's output only 17.15.48 # * JdGordon going to sleep 17.15.58 # no, the git import program 17.16.14 # pondlife: if you spot any other bugs, lemme know in the tracker, otherwise ill commit in the mrning 17.16.20 # luckily, git-svnimport doesn't 17.16.27 # :) 17.16.36 Nick JdGordon is now known as JdG|Zzzz (n=jonno@rockbox/developer/JdGordon) 17.17.17 # * GodEater 's fancy bash prompt is available to anyone who wants it - it looks best with a LC_ALL set to your_lang.utf8 17.17.53 Join dster [0] (i=dster@68-119-201-146.dhcp.spbg.sc.charter.com) 17.18.22 # and in a unicode terminal of course... 17.19.02 # damnit 17.19.19 # charger and usb pluged into my photo won't boot anything 17.19.29 # * Rincewind likes his prompt to be simple and small 17.19.44 # any other ideas to what i can try? 17.21.05 # dster: Leave it plugged in two hours, turn on hold, turn off hold, then hold down menu+select for 30 seconds. Try the menu+select a few times to make sure you're not letting your fingers slip slightly sideways 17.21.52 # Well, should I try to let it discharge before I try that? 17.21.54 # * GodEater is stunned 17.22.04 # a response from Mr. Torvalds himself! 17.22.05 # And, you mean plug it in the wall with hold off at first? 17.22.19 # GodEater: Was he reading our IRC logs? 17.22.45 # no - it's the git email list 17.22.50 # URL? 17.23.04 # I posted a couple earlier 17.23.29 # * linuxstb scrolls up 17.24.13 # http://article.gmane.org/gmane.comp.version-control.git/49106 17.24.18 # it's a nice response too :) 17.24.30 # damn the adresse is unreachable to m 17.24.35 # me* 17.25.14 # he didn't flame us :) 17.25.22 # GodEater: would you mind posting the response on pastebin ? 17.25.34 # Nico_P: there's another link further up not on gmane 17.25.59 # but sure : http://pastebin.ca/536912 17.26.26 # GodEater: His response was actually pretty much advocating how I imagined we'd best make use of git anyway. Heh. 17.27.54 # GodEater: can you post the link to the original message, too? 17.28.08 # man that's a long response 17.28.27 # indeed - he took his time... 17.28.43 # Llorean: indeed. 17.29.05 # I'm responding so he doesn't think we're idiots. (i.e. on the checking in binaries question) 17.29.10 # Hehehe 17.29.13 # Yeah 17.29.49 Join otih [0] (n=otih@p54A4F541.dip.t-dialin.net) 17.31.55 # what kind of binary files do we have? Just the bitmaps? 17.32.30 # Well, the original post referred to the actual build binaries. 17.32.35 # I think he thought we wanted to keep track of the rockbox bins 17.32.48 # But it wasn't talking about including them, just the fact that we need to build ones frequently from a central repository 17.33.54 # Rincewind: original message is also pastebin'd much earlier in today's log 17.34.23 # nvm, i searched around gmane and found it, thanks 17.35.57 # Llorean: yes, it looks like all of them interpreted the point about binaries wrong 17.37.37 # * GodEater hits send again 17.38.13 Quit atsea- (Remote closed the connection) 17.39.28 # * GodEater curses gmail once AGAIN for not doing reply-to-all by default. 17.40.19 # screw that, it should have reply-to-list 17.40.31 # my reply (for the impatient) http://pastebin.ca/536939 17.40.53 # Galois: yeah or that. I've lost count of the times I've replied to a list message, only to find I only replied to the poster, not the list. 17.41.05 # it makes me look like a complete moron (more so) 17.41.29 # not that that will surprise Mr. Torvalds at all I suspect :) 17.41.53 # maybe someone should go to google and give a talk on email 17.43.58 # GodEater: "I know you think everyone else is a moron" :D 17.44.17 # GodEater: maybe you could have done some top-posting to piss him off :) 17.44.22 # it makes sense if you watched the video 17.44.28 # I did 17.45.24 Quit linuxstb (Read error: 113 (No route to host)) 17.45.34 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb) 17.48.17 # markun: was very tempted ;) 17.48.18 Quit netmasta10bt (Read error: 104 (Connection reset by peer)) 17.48.37 # * GodEater is off to the gym. Later all 17.48.42 Quit GodEater ("CGI:IRC") 17.48.54 Join atsea- [0] (i=ariel@gateway/tor/x-e0663a7d450b7f76) 17.49.04 Join Nosgoth [0] (n=nosgoth@p54AADAEA.dip.t-dialin.net) 17.49.36 # Hi 17.49.51 # where can i ask for the wiki write permission? 17.51.52 # here 17.52.01 # wiki name? 17.53.16 # Nosgoth: what is your wiki name? 17.57.44 Join david1 [0] (n=david52@84.72.254.113) 17.58.25 # martintykal 17.59.14 # MartinTykal has write access 17.59.24 # thank you=) 17.59.25 Part Nosgoth 18.03.48 # austriancoder: the author of the wasabi stack is responding again :) 18.12.08 Quit Seed ("cu, Andre") 18.12.38 Join Seed [0] (n=ben@bzq-84-108-237-178.cablep.bezeqint.net) 18.20.53 Part Rick| 18.27.15 Join saratoga [0] (i=9803994d@gateway/web/cgi-irc/labb.contactor.se/x-ca7361498e3c6e07) 18.27.34 Join nls [0] (n=nils@nl104-202-175.student.uu.se) 18.27.57 # linuxstb: you around? 18.28.11 # I am. 18.28.57 Join Psilonaut [0] (i=DaPhunka@153.190-246-81.adsl-dyn.isp.belgacom.be) 18.29.26 # i am really confused with a problem 18.29.53 # I have tracked it down to the MDCT stage, where it occasionally spits out a pair of completely wrong values, generally evenly spaced about the center of the block 18.30.18 # however, I'm pretty sure nothing is underflow or overflowing 18.30.31 # is there some other hazard to look for? 18.31.03 # given that the fp version works correctly and the code is largely identical except for changing * to fixmul32 18.32.55 Quit lids (Remote closed the connection) 18.33.02 # I'm not the person to ask about the gritty details of codecs. Do you get the same problem in both the sim and on a target? 18.34.01 # sorry, this is just running from the command line 18.34.13 Join My_Sic [0] (n=MySic@mur31-1-82-237-204-133.fbx.proxad.net) 18.34.16 # on my linux machine, using a test driver 18.35.28 Join bluey- [0] (n=bluey@dslb-088-073-119-169.pools.arcor-ip.net) 18.35.56 # saratoga: you're supposed to work all summer on it, or did you plan a big vacation afterwards :p 18.36.15 Quit Rondom (Read error: 113 (No route to host)) 18.40.01 Join Lear [0] (i=chatzill@rockbox/developer/lear) 18.41.48 Join lids [0] (i=lds@gateway/tor/x-3212f6cd90ce5785) 18.42.31 Join Xerion [0] (n=xerion@cp198589-d.landg1.lb.home.nl) 18.42.45 Quit obo ("KVIrc 3.2.6 Anomalies http://www.kvirc.net/") 18.45.19 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") 18.45.52 # saratoga: Are you able to easily switch between using the floating-point MDCT and the fixed-point version? 18.47.49 *** Saving seen data "./dancer.seen" 18.52.08 Join Xerion_ [0] (n=xerion@cp198589-d.landg1.lb.home.nl) 18.54.51 # Lear: did you post your gcc hack anywhere? 18.55.09 # Nope. 18.56.03 # Lear: do you think it would be a good idea to use it for building rockbox or was it just an experiment? 18.57.33 # Well, it only makes gcc generate slightly better (and smaller) code for us, but it was more than an experiment, I'd say. Should probably file it as a bug for gcc 4.3 though... 18.59.21 # Lear: That might be a good idea, could you post it somewhere so I can take a look, I've never hacked at gcc but is a little curious :-) 19.00.05 Join _Xerion [0] (n=xerion@cp198589-d.landg1.lb.home.nl) 19.00.12 Quit Xerion (Read error: 104 (Connection reset by peer)) 19.00.23 Nick _Xerion is now known as Xerion (n=xerion@cp198589-d.landg1.lb.home.nl) 19.01.51 # I could send it to the developer mailing list, since the same "problem" might apply to ARM as well. 19.02.13 # works for me :-) 19.04.01 Join pepie34 [0] (n=pepie34@cop60-1-82-240-26-92.fbx.proxad.net) 19.07.08 # linuxstb: yes I've been going through and comparing them sample per sample to find bugs 19.08.08 Part Llorean 19.08.35 Join Caskey-10 [0] (n=Zach@75-133-92-85.dhcp.mdsn.wi.charter.com) 19.09.47 Join Caskey-91 [0] (n=Zach@75-133-92-85.dhcp.mdsn.wi.charter.com) 19.10.09 # hell? 19.10.13 # *Hello? 19.10.17 # hi 19.10.22 Quit Siltaar (Remote closed the connection) 19.10.35 # yea quick question about Rockboy on the 5g ipod 19.10.42 # 42 19.10.47 # (oh no :) 19.11.17 # rofl 19.11.28 # Caskey-91: well, let's get it over with.. 19.12.20 # * linuxstb thinks Caskey-91 was lying about the question being quick... 19.13.14 # maybe it was so quick that we didn't see it fly by 19.14.30 # Which reminds of Douglas Adams' quote regarding deadlines - "I love deadlines, especially the whooshing sound they make as they fly by" 19.14.37 Quit Xerion (" ") 19.14.49 # linuxstb: just like the 3.0 deadline :-D 19.15.32 # Precisely that sound ;) 19.17.30 Quit Xerion_ (Read error: 110 (Connection timed out)) 19.17.39 # linuxstb: you'll not believe it, but I have that quote hanging above our planning chart here at work ;) 19.18.01 # my boss didn't think it was funny 19.18.03 # petur: I thought it best not to mention it to my boss... 19.18.08 Quit davina_ ("xchat on Ubuntu 7.04") 19.18.32 # * petur is doing extra time at work :/ 19.21.38 Part david1 19.23.37 Join bluebrother [0] (i=uLJv2J5q@rockbox/staff/bluebrother) 19.24.32 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb) 19.25.16 Quit linuxstb (Nick collision from services.) 19.25.18 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb) 19.25.31 Quit aliask ("ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007060115]") 19.27.15 Quit Nibbier (Read error: 113 (No route to host)) 19.28.09 Join The-Compiler [0] (n=florian@233.49.76.83.cust.bluewin.ch) 19.29.29 Quit Caskey-10 (Read error: 110 (Connection timed out)) 19.29.33 Quit Caskey-91 (Read error: 110 (Connection timed out)) 19.30.50 Join Siltaar [0] (n=Siltaar@reverse-52.fdn.fr) 19.35.20 Join lids_ [0] (i=lds@gateway/tor/x-456c3db35bad1876) 19.35.35 Join desowin [0] (n=desowin@avc146.internetdsl.tpnet.pl) 19.35.40 Quit lids (Remote closed the connection) 19.38.41 Join obo [0] (n=obo@82-46-82-224.cable.ubr02.trow.blueyonder.co.uk) 19.38.52 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb) 19.39.00 Quit desowin (Client Quit) 19.45.47 Join Gwall [0] (n=Gwall@90.203.15.211) 19.46.55 Quit linuxstb (Read error: 113 (No route to host)) 19.47.23 Join Buschel [0] (n=AndreeBu@p54A3C9F7.dip.t-dialin.net) 19.51.54 Join Nibbier [0] (n=sven@e181076168.adsl.alicedsl.de) 19.58.15 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean) 19.58.38 Join linuxstb__ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net) 20.05.19 Join kaaloo [0] (n=luis@rue92-3-82-232-48-241.fbx.proxad.net) 20.05.26 Part kaaloo 20.07.16 Quit pepie34 ("Ex-Chat") 20.12.39 Quit linuxstb_ (Read error: 110 (Connection timed out)) 20.17.04 Quit Psilonaut (Read error: 104 (Connection reset by peer)) 20.18.14 Join bundik [0] (n=bundy@83-168-154-14.adsl.swan.sk) 20.20.20 Part bundik ("Leaving") 20.32.06 Join Rondom [0] (n=Rondom@p57A96110.dip.t-dialin.net) 20.32.45 Join kubiix [0] (n=Miranda@mos-81-27-201-28.karneval.cz) 20.34.01 Join GodEater [0] (n=vircuser@bb-87-80-121-64.ukonline.co.uk) 20.34.38 # woohoo - internet at home again at long long last 20.35.13 Quit petur ("work-> home") 20.35.36 # GodEater: Linus T's second response is actually quite good overall too. 20.35.39 Part Gwall 20.36.01 # yes, I read it ob the bus on my way home :) 20.36.06 Join DerPapst [0] (n=DerPapst@pD9EB3C13.dip0.t-ipconnect.de) 20.36.20 # * DerPapst is impressed 20.36.39 # I'm being thrown off the laptop though 20.36.42 # 2 replies by Linus T o.O 20.36.42 # be back in mo 20.36.45 Quit GodEater (Client Quit) 20.36.49 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn) 20.37.07 # and they were quiet nice too :D 20.39.14 Join ptw419 [0] (i=ptw419@216-188-249-122.dyn.grandenetworks.net) 20.43.04 Join GodEater [0] (n=bryan@bb-87-80-121-64.ukonline.co.uk) 20.43.39 # right back again 20.47.50 *** Saving seen data "./dancer.seen" 20.48.26 # DerPapst: yeah, he didn't call me a moron once :) 20.49.34 Quit orbit7 (Remote closed the connection) 20.54.01 Quit amiconn (Read error: 110 (Connection timed out)) 20.54.02 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn) 20.54.35 Quit Buschel (Read error: 104 (Connection reset by peer)) 20.54.49 Quit Mouser_X (Read error: 110 (Connection timed out)) 20.58.36 # my poor harddrive 20.59.05 Join maffe [0] (n=maffe@barmen.interhost.no) 20.59.06 # over a month since it last sync'd with portage - it's getting hammered 20.59.27 Join barrywardell [0] (n=barrywar@host-194-46-234-199.dsl-ie.utvinternet.net) 20.59.41 # heh 21.00.35 Join orbit7 [0] (n=martijn@cp562554-a.landg1.lb.home.nl) 21.01.49 Join Buschel [0] (n=AndreeBu@84.163.201.247) 21.01.51 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul) 21.03.06 Join mareviq [0] (i=magi@nat-mo1.aster.pl) 21.03.46 # GodEater: btw you called yourself a moron.. so he didn't had to do it anymore :P 21.03.56 # GodEater: I managed to run VB .NET apps in linux :) 21.04.11 # not all of them and not the main one I wanted, but still 21.04.48 # the mono VB compiler seems very picky though, so I didn't manage to compile any programs 21.05.47 # v. impressive Nico_P :) 21.06.30 # DerPapst: I didn't think I called myself that - I think I just commented that he though everyone else was a moron. Not *quite* the same - but I see what you mean :) 21.06.36 # I'm impressed by mono... very nice project 21.06.57 # and congrats on not being called a moron by LinusT :) 21.10.27 # I've been plating with git-svn and I'm still finding it a bit hard 21.10.38 # s/plating/playing 21.10.54 # Well, as was said, if we move to git, it's going to be all or nothing from the sounds of it. 21.12.29 # well actually from what I understood, there's nothing preventing some of us from having git repos mirroring the SVN repo for a while, and pushing/pulling from each other 21.12.55 # it's far from optimal of course but it could work 21.13.34 # yep, there's nothing anyone can do to stop people with commit access from using svn-git 21.13.44 # Indeed 21.13.58 # It might be a good way to both learn git some more, and get used to using it for better collaboration. 21.14.19 # It might even be good for the SoC stuff, making it quite easy for the mentor to pay attention to the work being done by the student. ;) 21.14.32 # yes, there could be some separate git repos for experimenting 21.14.58 # as I said, I think dan_a already uses git-svn 21.15.00 # Llorean: I was thinking of a SVN branch for that, but git does sound appealing :) 21.15.18 # Nico_P: git would certainly make it easier, I think 21.15.46 # btw you might be interested in knowing that I've been thinking about my project but haven't had time to actually start on it 21.16.08 # I was going to ask where you're standing right now. ;) 21.16.21 # my shcoolyear isn't finished yet and I have a few exams left 21.16.27 # Gotcha 21.17.20 # and I'm kinda maturing the ideas in my head, probably a bit like amiconn does with viewports :) 21.17.55 # hopefully there'll just be less time between the idea and the code 21.18.33 # Sounds good 21.19.22 # I've been thinking about pondlife's BufferingApiProposal and discussing it with him a bit 21.26.48 # I've read a few logs of that in channel at least 21.27.27 Quit Nibbier (Read error: 110 (Connection timed out)) 21.36.07 Quit The-Compiler (Remote closed the connection) 21.36.48 Nick linuxstb__ is now known as linuxstb (n=linuxstb@i-83-67-212-170.freedom2surf.net) 21.41.09 Join perpleXa [0] (i=perpleXa@unaffiliated/perplexa) 21.41.37 Join Xerion [0] (n=xerion@cp198589-d.landg1.lb.home.nl) 21.43.08 # linuxstb: tried to download the demac tarball, but it ends up corrupted here. Don't know yet if it is a simple cr/lf issue... 21.44.59 Join Faemir [0] (n=daniel@85-211-222-38.dyn.gotadsl.co.uk) 21.46.00 Quit tchan (Read error: 110 (Connection timed out)) 21.46.38 # linuxstb: He, had to run 'gzip -d' first, then 'tar xzf' worked... 21.46.49 # Lear: Yes, I've just noticed that myself... 21.47.40 # I've had problems in the past with gzipped attachments on flyspray being gzipped twice. The file I uploaded was definitely only gzipped once... 21.47.56 Join davina [0] (n=dave@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com) 21.48.04 # Are you using Firefox? 21.48.52 # If I download the file with wget, it's fine... 21.49.04 # With firefox, I end up with the double gzipping effect... 21.49.39 Join kaaloo [0] (n=luis@rue92-3-82-232-48-241.fbx.proxad.net) 21.50.01 # Firefox, yes. 21.51.00 # is it on purpose that emacs is ugly ? 21.51.27 # I just built v22.1 with hopes that the GTK interface would like nice, but it doesn't at all 21.53.13 # OK, looks like GTK has to be activated in the configure script. I'm shutting up and rebuilding 21.53.24 Part kaaloo 22.00.46 Quit bluey- ("Leaving") 22.02.25 Join BRi7Xxon [0] (i=bri7x@c-68-81-211-195.hsd1.pa.comcast.net) 22.02.39 Quit alienbiker99 (Read error: 104 (Connection reset by peer)) 22.02.54 Quit BRi7X (Read error: 104 (Connection reset by peer)) 22.04.02 Join alienbiker99 [0] (n=alienbik@ool-44c126d4.dyn.optonline.net) 22.07.44 Join hostf4cekilla [0] (n=chatzill@pool-70-22-204-247.bos.east.verizon.net) 22.07.52 # Lear: just rebuilt gcc with your "hack", everything seems to work nicely, some quick tests with test_codec shows no speed difference at least for vorbis. 22.08.24 # No surprise there (about the speed difference). 22.08.57 # quite a few of the codecs also dropped a few hundred butes in size 22.09.04 # bytes 22.09.05 # Lear: I think that if we want to use that patch, we should just increase the "rockbox patch number" 22.09.19 # I.e. this would become rockbox patch #2 22.10.08 Join TrueJournals [0] (n=aimjourn@c-24-12-147-61.hsd1.il.comcast.net) 22.10.20 # Does this also work for 4.0.x? 22.10.29 # Makes sense. But it would be a little confusing with #1 being for 4.0, and #2 for 3.4... 22.10.33 # * amiconn would like to switch to 4.0.x for coldfire as well 22.10.44 # * nls would like rockboxdev.sh to handle the patch for building m68k-gcc on amd64 too 22.10.45 # Haven't looked, but I'd guess it would be similar, if not identical. 22.11.02 # Most things run at the same speed, and wavpack decoding becomes faster 22.11.27 # Didn't try any encoder though... 22.11.58 # No, same patch wouldn't work, but manual modification is simple. 22.13.44 Part mareviq ("nie wszystko dioda co sie swieci...") 22.15.14 # is there any reason that keeps us from switching to gcc4 for coldfire? 22.16.31 # Slower code, I guess... 22.17.27 Quit ompaul ("reboot, reconfigure, retest, hope") 22.18.45 Quit _Veseliq_ (Remote closed the connection) 22.19.45 Join limbus [0] (n=limbus@port-83-236-56-26.dynamic.qsc.de) 22.20.00 # hi all 22.20.53 # GodEater: the mail was a success =) 22.21.08 # why are we only building for the archoses with -Os and just -O for others? shouldn't -Os be both faster and smaller than -O? 22.24.04 Join jaczehack [0] (n=rockbox@c-b7f772d5.013-1-73746f40.cust.bredbandsbolaget.se) 22.24.53 Quit GodEater (Read error: 113 (No route to host)) 22.27.01 Join GodEater [0] (n=bryan@bb-87-80-121-64.ukonline.co.uk) 22.30.21 Quit kubiix ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org") 22.31.56 Join din [0] (n=din@c-67-162-206-27.hsd1.ar.comcast.net) 22.32.34 Quit rift ("d'accord") 22.34.47 Quit Thundercloud (Read error: 104 (Connection reset by peer)) 22.35.14 # nls: Try -O2 or -Os ... 22.35.15 Join petur [0] (n=petur@rockbox/developer/petur) 22.35.25 # * amiconn expects crashes and/or failing builds 22.35.56 Quit jaczehack (Remote closed the connection) 22.36.06 Join Thundercloud [0] (n=thunderc@82.152.213.103) 22.36.23 # amiconn: ah, ok, I have my last exam tomorrow, after that I will play around with it, thanks for answering my questions :-) 22.37.02 # It was quite a bit of work to get the archos builds actually work with -Os (and it is the reason for rockbox patch #1, fixing a gcc bug) 22.39.34 # sorry for the silly question (I am not use to linux) but where I modify the path in linux, I got this message : The compiler you must use (arm-elf-gcc) is not in your path! 22.40.40 Join barrywardell_ [0] (n=barrywar@host-194-46-234-199.dsl-ie.utvinternet.net) 22.42.03 # toffe82, IIRC, you got to edit (as root) the file /etc/profile 22.42.04 # toffe82: you can change it in your .profile file /home/youruser/.profile 22.42.13 # you'll find PATH=... 22.43.21 # limbus: I don't use linux but unix, doesn't linux have an independent .profile for every user? 22.43.40 # I think one overloads the other 22.43.53 # I'm just looking, but not finding either 22.44.00 # YOu can extend the path just for your user in the user specific profile 22.44.01 # use ls -la 22.44.14 # the dot at the begining makes it "invisible" :p 22.44.24 # It depends on the shell, for bash (which is standard on most distros) it's .bashrc 22.44.45 # Just put a line in there reading like: 22.44.54 # I am on ubuntu 22.45.04 # export PATH=/path/to/your/arm-elf-gcc:$PATH 22.45.08 # ls 22.45.13 # sorry ;) 22.45.23 # :p 22.45.39 # toffe82, try "ls -a" or files starting with "." will not be shown 22.46.08 # Then log off and log on again to activate the change, or execute the same command directly on the commandline 22.46.38 # or run ". ./.profile" 22.46.38 Quit Xerion (Read error: 104 (Connection reset by peer)) 22.47.07 Join tchan [0] (n=tchan@lunar-linux/developer/tchan) 22.47.09 # ". ./.bashrc" if you're running bash shell (per amiconn) 22.47.16 # in nautilus, press CTRL-H to view hidden files 22.47.54 *** Saving seen data "./dancer.seen" 22.48.27 Quit barrywardell (Read error: 113 (No route to host)) 22.48.32 Quit petur ("switch!") 22.48.49 Join petur [0] (n=petur@rockbox/developer/petur) 22.50.25 Quit saratoga ("CGI:IRC (Ping timeout)") 22.52.29 # amiconn: patch is now on CrossCompiler wiki. 22.52.59 # I want to add my server (at home, SDSL connection) to the rockbox build farm. rockboxdev.sh just finished, and I think I am ready to receive the public key and acbuild.pl 22.53.36 # mayber Bagder can guide me through the rest of the process ? 22.53.50 Join Rob2222 [0] (n=Miranda@p54B14982.dip.t-dialin.net) 22.53.57 # he's listed on the BuildServer wiki page 22.54.19 Nick Caliban_ is now known as Caliban (n=ianmacd@kwark.caliban.org) 22.54.41 # is it a fast machine? 22.54.52 # Intel Core 2 Duo 1,86 22.55.06 # 1 GB ram 22.55.15 # http://daniel.haxx.se/rockbox/acbuild.txt is the script that should be named acbuild.pl and get +rx permissions 22.55.24 # ack 22.57.48 # * petur kicks austriancoder awake 22.58.02 # limbus: check the PATH at the top of acbuild.pl so that they match your system. you have sdl installed too? then tell me your host name and I can do a first simple verification 22.58.22 # I'll tell you as soon as the public key is in place 22.58.23 # petur: ping 22.58.28 Part TrueJournals 22.58.43 # does rockboxdev.sh check for sdl ? 22.58.47 # nope 22.58.55 # try building a sim 22.59.01 # austriancoder: http://users.telenet.be/petur/usbotg20070704.zip 22.59.02 # jep 22.59.31 # petur: thanks 22.59.57 # austriancoder: warning, the code builds but is a mess 23.00.21 # and the author of that code just returned my mail today 23.01.32 Quit Lear ("ChatZilla 0.9.78.1 [Firefox 2.0.0.4/2007051502]") 23.01.35 # petur: will start looking into a nice layer to access different usb chips... 23.01.47 # great 23.03.06 Quit pondlife ("disconnected has pondlife") 23.03.54 # just curious...how's the work on the usb driver going? 23.04.07 # evaluating stacks 23.04.48 # and as a sideproject, I'm trying to get the usbotg driver running on h300 23.04.54 # petur: we will need more different layers i think. one for device part, one for host/otg part.. what do you think? 23.05.24 # forget otg, just host and device 23.06.07 # I need to look more at that code to see how it switches 23.06.09 # I looked at the lpc214x usb stack in the past 23.06.45 # barrywardell_: there's a wiki page about the stacks austriancoder evaluated 23.06.58 # * barrywardell_ checks it out 23.07.17 # http://www.rockbox.org/twiki/bin/view/Main/UsbSoftwareStack 23.07.39 # thanks 23.09.47 Join pepie34 [0] (n=pepie34@cop60-1-82-240-26-92.fbx.proxad.net) 23.10.19 # petur: md5sum okay? 8708f314c6849842190bd69a6058f757 usbotg20070704.zip 23.10.27 # I have problems to open the zip 23.11.26 # I'll email it, the webspace is probably full 23.11.47 Quit davina (Remote closed the connection) 23.11.49 # fine 23.11.51 Join entheh [0] (n=purr@88-106-199-196.dynamic.dsl.as9105.com) 23.11.54 Part perpleXa 23.13.32 # austriancoder: check your gmail box 23.14.18 # petur: thanks.. now it works 23.15.03 # * petur has too many devcon pics on his webspace... damn ISP giving so little 23.17.20 Quit Faemir ("I could put something witty here, but no...") 23.20.09 Quit Thundercloud (Remote closed the connection) 23.22.38 Quit Rondom ("Ex-Chat") 23.24.22 Quit pepie34 ("Ex-Chat") 23.26.47 Part nls 23.27.21 Join miepchen^schlaf [0] (n=hihi@p54BF51C1.dip.t-dialin.net) 23.27.23 Quit Siltaar (Remote closed the connection) 23.30.36 # so, anyone feels like doing a commit? ;-) 23.30.50 # give me 10 minutes ;) 23.30.54 # :) 23.31.07 # you can have 8, no more 23.31.10 # :-P 23.31.26 # then he'll do in 6 :p 23.31.59 # * petur slaps Bagder with a large something 23.32.18 # * joshin knows that 10 minutes to a programmer is twice as long as a minute is to a wife/girlfriend getting ready to go out 23.32.46 # lol 23.32.47 # I think it's more like a factor pi 23.35.36 Quit jgarvey ("Leaving") 23.35.50 Join Soap [0] (n=Soap@rockbox/staff/soap) 23.38.43 # done 23.38.53 # in 8 :) 23.39.48 # haha 23.39.57 # n1 23.41.17 # wow... are we down to 4 minutes for all builds? 23.42.58 # indeed impressibe 23.43.03 # uh impressive 23.45.04 # looks like it is hanging somewhere? 23.45.18 # yeah, godeater... 23.45.29 # "since 348" 23.46.43 Join Cardini [0] (i=189fceb0@gateway/web/cgi-irc/labb.contactor.se/x-f2e4e1c6574a776f) 23.47.14 # one of these days I need to write a fix for that 23.48.07 # restart ? 23.48.12 # yeah 23.48.24 # I don't have any good way to just nuke one single server's works 23.48.24 # I have a problem: my player is stuck on the RockBox start-up screen, with the dialogue "Committing database [1/8]." This seems to be because I chose the "Initialize Now" option in General Settings>Database. Model is Sansa e270 6GB, version is current. 23.48.39 # heh, you're going to rewrite godeater? Good luck :p 23.49.00 # can't be that hard, can it? ;-P 23.49.03 # lol 23.49.27 # Bagder, cpu-load still hilarious (less 10%) 23.50.34 # limbus: are you doing "make -j n" ? 23.50.34 # Cardini: How long has it been there, and how many files do you have? 23.51.09 # Nico_P, -j 2 on a Intel Core 2 Duo 23.51.26 # not sure whether I can set 4 23.51.29 # ant you only have 10% CPU load ? 23.51.34 # s/ant/and 23.51.41 # yes 23.51.41 # grrr, after 190 seconds(!) one build is left... on godeater 23.51.52 # eat 'em ! 23.51.55 # :) 23.51.57 # can't you just diable him ? 23.52.14 # rrrr... I need to learn to type 23.52.19 # I need to, yes 23.52.19 Quit orbit7 (Read error: 104 (Connection reset by peer)) 23.52.23 # Llorean: It was up for 15 minutes before I reset the player, and additional 21 minutes after restarting. I then reset it again and am charging it using the original firmware. Files? About 550 music files, and I think the total is about 920. 23.52.40 # I wish I knew why it happens... 23.53.31 # Nico_P, precisely "-j2" whitout spaces 23.53.50 # ccache enabled 23.54.26 # * amiconn whispers "speculative parallel builds" 23.54.38 # 24% 23.54.43 # on one of the cpus 23.54.50 Quit miepchen^schlaf (Read error: 110 (Connection timed out)) 23.54.52 # amiconn: that's not really the answer to this problem though 23.55.11 # in fact 23.55.15 # No, but it reduces the severity problem 23.55.16 # auto-kill after xx time? 23.55.21 # it makes this problem harder 23.55.25 # Insert "of the" somewhere.. 23.55.41 # yeah, I need to kill jobs taking more than N secs 23.55.54 # why? 23.56.05 # because they're dead 23.56.34 # and if we parallel build, another server will build and provide a zip 23.56.43 # or might at least 23.57.13 # The idea of speculative parallel builds is that in the end, several servers are building in parallel. When the first server returns the finished build, all parallel versions of that build would be killed 23.57.24 # yes of course 23.57.27 Join Xerion [0] (n=xerion@cp198589-d.landg1.lb.home.nl) 23.57.37 # but we can't be sure that will always happen 23.57.44 # 209 seconds 23.57.52 # that's pretty cool 23.57.53 # whoopsie: limbus.dyndns.org 29 0 0 ( 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0) 23.58.00 # haha 23.58.03 # that's why then 23.58.05 # that explains the cpu :) 23.58.16 # So as long as the whole build system doesn't break down to a single server in total, even a hanging one shouldn't cause problems 23.58.43 # amiconn: unless there are several builds that hang at once 23.58.44 # * petur hands 580 points to limbus 23.58.49 # :)