From 965058028f08153fe4920afcd0c835297bd5c68d Mon Sep 17 00:00:00 2001 From: Daniel Stenberg Date: Sat, 20 Apr 2002 14:23:46 +0000 Subject: four new logs git-svn-id: svn://svn.rockbox.org/rockbox/trunk@152 a1c6a512-1295-4272-9138-f99709370657 --- www/irc/rockbox-20020418.log | 559 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 559 insertions(+) create mode 100644 www/irc/rockbox-20020418.log (limited to 'www/irc/rockbox-20020418.log') diff --git a/www/irc/rockbox-20020418.log b/www/irc/rockbox-20020418.log new file mode 100644 index 0000000000..4ae09bf7f6 --- /dev/null +++ b/www/irc/rockbox-20020418.log @@ -0,0 +1,559 @@ +**** BEGIN LOGGING AT Thu Apr 18 08:42:39 2002 + +--> Bagder (~daniel@as3-3-2.ras.s.bonet.se) has joined #rockbox +--- Topic for #rockbox is Open Source Jukebox Firmware - http://bjorn.haxx.se/rockbox/ +--- Topic for #rockbox set by Zagor at Fri Apr 12 15:45:52 +--> Zagor2 (~bjst@labb.contactor.se) has joined #rockbox +--- Zagor2 is now known as Zagor +--> Linus (~linus@labb.contactor.se) has joined #rockbox + wtf is the story with the makefile? + there's a story? + umm.. well.. all the >>>>> and <<<<<< and ======= stuff + ah, that's a conflict + you and someone else changed the makefile + the >>>>>, <<<< and ===== describe the two versions + which makefile btw? + uisimulator + nevermind.. i just did and update on my end... + ok +adiamas adi|atWork could someone tell me where their lcd.h is located? + i don't have one and i need to grab it off a remote box... + makefile is complaining + ah + it's in the firmware module + I have uisimulator and firmware checked out in the same dir + ? + so in the uisim dir I can get it ../firmware/lcd.[ch] + im really starting to dislike cvs + rockbox, rockbox/uisimulator, rockbox/firmware + yeah.. but _i_ don't wanna work on firmware + you do now ;-) + but your telling me that i need to pull it down and keep it up to date huh ? + no but you need that source. we don't want to duplicate it + because the uisim uses firmware code +* adiamas giggles + can't we break it out into another "library" or "common" dir or something? + and have a third module you need? + seems that makes more sense.. + nope.. still only be two.. + why should i worry about firmware if i only want to play with the uisim + thing is, I want the uisim to use as much "real" code as possible + nods + because the uisim simulates the firmware + nods + okay.. now.. because im stupid.. if i just want to pull the files down and not checkout... + whats the cvs command? + or do i co no matter what? + yes + there's only checkout, yes + nod + im so much more used to sccs :) + but since you don't modify them, you won't get problems updating them in the future + sccs is good, but not for widely distributed development +* adiamas nods + i understand the use of cvs.. just feels icky cause im not used to it + widely/wildly + it takes some getting-used-to, I agree +--> wavey (~wavey@dlan1431.dircon.co.uk) has joined #rockbox + hey ho wavey +--- Zagor is now known as Zagor|work + goodly morning + hmm... okay.. anyone know wtf firmware/system.h would bitch on uisim compile? + complaing about the #define \'s + hm + lemme check + what does it say? + it works here + gcc -DHAVE_CONFIG_H -DGETTIMEOFDAY_TWO_ARGS -DSIMULATOR -DHAVE_LCD_BITMAP -I../firmware -g -c uibasic.c + In file included from ../firmware/lcd.h:24, + from lcd-x11.h:24, + from uibasic.c:36: + ../firmware/system.h:52: parse error before `volatile' + ../firmware/system.h:62: stray '\' in program + and then like 100 iother stray complaints.. but i figure those are ghosts + what gcc version is this? +* Bagder suspects the crlf newlines... + gcc version 2.95.3 20010315 (release) + im not switching to the 3.0 series till its aged a bit + on linux, right? + yup + ill tell you if its crlf + one sec + looks like it... + i load it in emacs and it comes in DOS format + mode rather + yes + so try stripping them off and compile and see if that removes the warnings + ummm.. how do i do that easily? + tr -d '\015' < system.h >system2.h + then use the version 2 of course + yup.. that did it... + ok + so we need to be more careful on the newlines + so when i get diffs on what i do and others do.. clashes and what not.. and we get the <<<<<<<<<< and> >>>>>>> complaints.. how do we resolve those? + the makefile can include a newline checker +adiamas adi|atWork adiamas adi|atWork thats a good idea + adiamas: those are conflict markes that appear when you get an update that has changes on the same lines you have changed locally + right.. i know.. but how do we resolve them... + ie: you change int i to int duck + and i want int wolf + what happens? + you edit out the "wrong" piece + and keep the correct + hehe but which is wrong? + you need to figure that out + it's a conflict with your changes + we can't tell you what's right ;-) + but it may only be a conflict with my changes because someone elses bug fix may be diff then mine.. and it clear out theres and place mine.. then theirs is in conflict + and round the wheel turns... + yes, but that's why you need to pick one of the alternatives. Or write a new approach ;-) +* adiamas reminds you he's still new to this ;) + one last question, then bed... is there a way to set my searchable include dirs in my env so that the makefile can find what it needs if things are in diff locations? + it's possible to pass as arguments to make + bah.. was trying to avoid that... + it could be environment variables too + we should figure out a way to let people do this easily +* adiamas nods + fixed.. settled for a symlink and said screw it + way to go ;-) + adiamas: -IX -IX+1 -IX+2 for your makefile include entry will search X, X+1 and X+2 .. + where X, X+1 and X+2 are include dirs + if i understand your q correctly + so people edit their makefile to include all their includes in the $(INCLUDE) directive + that's how I do it + especially on cygwin, where the includes are all over the place + oh, and I replace the use of -I$(INCLUDES) with simply $(INCLUDES) + to ensure the first -I doesn't translate into -I-IX +* wavey goes back to OO modelling + of course, we could always use autoconf +* wavey runs + hehe + wavey nah.. that i know.. the part that was issue was the libs.. + yeah.. editing the makefile isn't an issue.. + i just don't want to have to keep fixing it each time i pulled down a clean copy and all + so i just tossed LIBDIRS in the env + and do make -e +* adiamas shrugs + oh, libs - i thought you said includes + hehe nope... + im used to makefiles.. just not used to dealing with that on a new starting project :) + w/ other ppls machines and all + 'is there a way to set my searchable include dirs in my env' + ha +* wavey knew he wasn't going mad + hehe sorry... + i think i need bed too + 1 - 0 to wavey ;-) + but i've been awake only a couple hours + its 0430 here.. cut me some slack ;) + heh + go + to + bed :) + can't sleep :( + this heatwave here is killing me.. + came to quick... + where are you? + NY state, usa + bout, 30 mins north of NYC + aha + how hot can it be at 0430? + hit 93 today at work... + ouch + yeah... + hope the ac is working + don't have any + just a ceiling fan + double ouch + nods + i find turning lights off and working by candle makes it cooler (both definitons of cooler) +* wavey blinks + besides.. the gf comes up this weekend.. so sweat is a good thing then ;) + heh + btw.. its no ac at home.. work has it.. + ah + glad to hear it +* Bagder counts, 33.9 celsius = 93 fahrenheit = 307 kelvin + uk has been mild/warm for a few weeks now + s'lovely + had the first rain in ages yesterday + we have like 9C today... :-/ + hehe we are in a near drought situation already... no lawn watering, town just passed restrictions on pool filling + wow + and across america over 100 records were broken due to the high heat + we use CDs in the uk now + NYC had it's hotest temp on this day in history, for the 3rd day in a row. + they're smaller and shinyier +* adiamas quirks his head to the side and looks at wavey + i don't get it. + records broken + cd + ah.. + heheh + it's late + you're tired + im ugly too :) + its understandable :) + heheh + but shhh.. thats a secret +* Bagder looks at wavey, oooooh that was a pretty bad one ;-) + everyone's beautiful in ascii + lol... i like that... + congrats.. you just joined my sig file :) + eheh + heh + in ebcdic however, everyone's a pig + is that why we've switched to ascii these days then? B-] + yup :) + and frosted mirrors too + wavey.. whats your real name so i can credit you all proper like :) + stuart martin, if you need the credit. take it for yourself with my pleasure + btw.. its 70 F here, 21 c + UV Index: + 0 Minimal + heheh + i wonder why... + could it be the suns not up? + :) + when is dawn these days? + any of you read scifi? + dawn as in sun rise or dawn as in the book? + sunrise in NY 05:12 + almost finished two olaf stapledon books. 1930s author - one of his books describes the history of humanity (over 200 000 000 years) and the other describes the history of the universes + both fantastic, btw + nods + (according to my psion) + absolutely recommend them + wavey, feel like mailing me the book names? + either that.. or msg me them... + 'last and first men' + and 'starmaker' + nods + the channel is quiet, i'm sure no one cares :) + got it. + okay.. bed time + night all + night adiamas + see ya +--- adiamas is now known as adi|sleep +* Bagder is curious on Linus I2C and MAS work... + i2c? + whassat + ooh, only 15*6 songs left on my random playlist of 3500 songs :) + that's how to speak to the MAS circuit + one quick note.. did you guys realize there is a key lock on the jbr? + yus + on + nods + hehe i din't :) + back to bed :) + heh + night +* Bagder has hardly used his recorder yet :-) + we need a tiny math lib for our code + we do + which projects can we steal it from? + I dunno actually + we could probably "steal" function by function + from BSD or Linux code + sure + be nice if sourceforge allowed a search through open source project's codebases for things + be very useful + true + make it easier for industry to rip it off as well i guess + not that it doesn't happen already anyway + The I2C is going slow at the moment. But the MAS is responding. + *cool* + ...occasionally. :-) + hah + occasionally cool :-) + heh + what about data structures? + we cutting our own linked lists, etc? + we could probably use a sensible single implementation for many to use + must be a nice and small and tested-to-hell one in apache or somewhere + yus +--- Zagor|work is now known as Zagor + btw, we're having a litte "rockbox" gathering tomorrow + the stockholm/sweden parts of the rockboxers + how close do you guys live together? + 20km + 'variable pitch control' is that achieveable by the MAS? + (from the mail list) + nope + can't be done + rockbox lunch time + hey you know if the DAC can have 2 channels on at once? + I don't think so. Lemme check. + It seems like we can mix all channels, AUX1 AUX2 and DAC + Cool. Then we can hear the RS232 data when we are debugging. :-) +--- Zagor is now known as Zagor|lunch + yeah? nice +--- Zagor|lunch is now known as Zagor +--- wavey is now known as wav_lunch +* wav_lunch smiles + timezones. who'd have 'em? + they're useless +* Zagor wants to do away with timezones and daylight savings time + yus + one hour, globally + who cares if 0515 is in your afternoon? + exactly. it's just a bloody number +* PsycoXul agrees + swatch tried it with Internet Time + only met one person with such a watch though + if the world's consumers had embraced the concept, the governments may have agreed to it + anyhow + lunch + the problem was that they wanted to alter everyone's sense of time. the changed too much + people still want 12+12 hour days, 60 hours minutes etc +--> alkorr (jbcoax@srs04m-2-107.n.club-internet.fr) has joined #rockbox + Welcome Alan! + hi linus + hi there + i have a question + Ask! + quite now we cannot use other formats than MP3 + not even a WAV file + the recorder is able to get a linear PCM on digital line + True + if i'm not wrong, wav format is a PCM-like stream + yes + wav is usualy just pcm with a header afaik + my idea is the following + if we strip unused infos + and feed internally the digital pin of MAS + with a PCM stream (using sci0 ?) + and set MAS for recording + would you hear this wav playing ? + I don't follow you. Do you want to play WAV files on the recorder? + a trick which could allow recorder to play wav files (not forgetting possibility of mixing) + i read on the company's website about the player's MAS and they claimed to have pcm playback firmware for that chip anyways, likely the recorder's chip too... of course getting any such things i dunno, never found any downloads etc + The MAS is capable of playing WAV files. You just have to download firmware to it. + i think it is not free + can archos firmware do whats necessary to download firmware to the MAS? + i tried to get this addition but without success + yes it could + if it have additions to upload + has + but i'm doubtful that Archos will buy this addition for its player/recorder. + i think we need a dsp hacker to the project :) + of course if we could know what kind of DSP it is and have compilers, it would be great + yes + but Micronas are unlikely to reveal anything about this DSP + why? + Do you know a lot of company which reveal all the works ? + their + what DSP they use is hardly "all the works" + anyway, companies are not bound for revealing something :) + perhaps they don't want to let a non Micronas employee to be able to reverse-engineer their works + they are commercial, not open free :) + well + they DO have a devkit that they supply to their customers + they mention it and some of the things in it + example firmware and such + yes but if they don't let people write dsp code, who will buy their dsp? + customers => archos, not final users + customers must pay for it ! + true + including some voice compression codec decoder and encoder, and pcm player + yeah + but there are a few customers of the MAS + perhaps we could ask around + so somebody find a friend of a friend of a friend who works for a customer ... + heh + okay if you can a smart guy to help for retrieving missing infos and even get tools, that would be great + find + getting the tools is probably not legal + and if he/she is not concerned with a NDA too + I doubt all customers sign NDAs to buy the MAS devkit + ya, if we could have opcodes, we can create an assembler + but what i know it should be 24-bit opcodes (code addresses are 24-bit) + btw, if you look at MAS commands, to upload a firmware you must issue a command and a block of 32-bit words + but only the first 24 bits in those words are valid + (or it should be 12-bit ?) + hum found a 1 MB pdf... + the data sheets speaks of 20-bit words + ok + so it is 20-bit + i found a page where someone said what DSP was in fact + wow + i try to get it back + MASC + 3500 ? + not 35000 ? + anyway not a target we can find for gcc or as :( + uh no + http://winwww.rutgers.edu/~samar/platforms/masc3500.html + ya + found + nice digging! + the guy who wrote that web page could have some info + yes i tried but i wonder if his email address is still valid + he has two, try both ;-) + hum i prefer let you to try both ;) + "MASC 3500 is NOT supported by C compiler. Only assembler programming is available" + actually, I think I'll hand over that work to Björn + okay + *thump* + got it + :) + he's the man + ;-) + :) + so..um... what do we want to know, exactly? + whether he has an archos for a start :) +--- wav_lunch is now known as wavey + have all the information he can feed us about this DSP : especially all the opcodes + haha + haha + "I was especially interested in digital modulation techniques and speech processing. So, embedded systems and numerous real-time DSP implementations (telephone line modems, low bit-rate speech coding...)" + he's our man! + There was only one commercially available MPEG layer 3 chip, the MASC3500 from ITT Intermetall Semiconductor. We attempted to obtain information and availability of the chip, but the company is situated in Germany and gave very little feedback + not a good sign + yes it is the main culpritt of thy company + this + humm i mean the main discredit for this company :) + your point is usually clear, al :) + A. Domazetovic, D. Samardzija, "Implementation of 4880 bps CELP on MASC 3500 DSP", Conference on Telecommunications ETRAN, Vrnjacka Banja, Yugoslavia, Jun 1998. + anyone near a library? + in yugoslavia? +* wavey chuckles + lol + D. Samardzija, Z. Lukac, "A Solution of CELP Encoder at 5.3kbit/s and 6.3kbit/s on ITT MASC 3500 Processor", Conference on Telecommunications ETRAN, Vrnjacka Banja, Yugoslavia, Jun 1998. + well in a very good english : "this lack of feedback is the rock(box) on which the undertaking comes to grief" ;) + yay, we've got a poet! ;) + al: and now in italian, please :) + italian you are ? + (without using babelfish) + italian I am not + english I am + arf ;) + arf is a lovely word :) + "questa mancanza di risposte è la roccia(boccia) su cui l'impresa viene al dolore " ;) + roccia boccia +* Bagder giggles + arf => lol + fantastic! :) + roccia boccia is mighty + huh "boccia" doesn't exist :P + Wav: found something about "A Solution of CELP Encoder at 5.3kbit/s and 6.3kbit/s on ITT MASC 3500 Processor" ? + it's one of Dragan's publications + okay so no pdf at all ? + no + i'll email him and see if he's a cool guy or not + hmm, this guy wrote a CELP encoder for the MAS 3503, which is the predecessor of the 3507 we have in the jukebox + i think i'm in love + arf :) +* Bagder worships Dragan a little + you think it helps? ;-) + Why ? + brb +**** ENDING LOGGING AT Thu Apr 18 14:44:53 2002 + +**** BEGIN LOGGING AT Thu Apr 18 15:27:51 2002 + +--> Bagder (~daniel@as3-3-2.ras.s.bonet.se) has joined #rockbox +--- Topic for #rockbox is Open Source Jukebox Firmware - http://bjorn.haxx.se/rockbox/ +--- Topic for #rockbox set by Zagor at Fri Apr 12 15:45:52 + does wordexp() exist in windows? + anyone? +* Zagor doesn't do windows + wordexp ? + Bagder: I just managed to read the version number from the MAS via I2C! Yippie! + *neat* + wordexp, yes, it's a posix function + though I doubt windows features it + use the cygwin.dll from Cygwin ;P + i'm doubtful too + unless they have a posix library + what is its purpose + "perform word expansions" + i cannot find its manual entry + heh, no I couldn't either on linux + really a standard posix or just a mandatory extension ? + then Linux man pages contain many holes + standard posix + single unix specification + never mind + okay + so what is the version number of the MAS? ;-) + Linus: player or recorder ? + it's the player + actually, he read the model number so it's "3507" + woooo + we didn't know that! B-P + hehe + Linus: you should try I²Cize your MAS on the recorder too + maybe he faked it :) + alan: yeah, it's just a lot easier to do development on the player since he's got the serial port on it (and the I2C wires connected to a digital oscilloscope...) + ah yes, the wires... + you guys are gonna make these little boxes do things the archos people never dreamed :p + just thought i'd throw that in there before running off to bed heh + ... and not some things people do dream of (judging from some of the feature requests) ;-) + heh + yeah well + My I2C test was on the player + Recorder has two I²C devices so don't be surprise if you code could not work ;) + Bagder: that may not be all bad heh + and theres a lot of requested features that would be like.. moot with new better firmware + you know things that apply to the current firmware's limitations etc + that's true + Why? I assume they have different ID:s. + especially, if you requeste a RTC reading via an interrupt while reading something on MAS ;) + Hehe. Who would do such a thing? :-) + They do + :-) + anyways, sleep time.. night + night PsycoXul + night + nacht Psycho + hum i must leave, see you ! +<-- alkorr has quit () +<-- Linus (~linus@labb.contactor.se) has left #rockbox +<-- Zagor (~bjst@labb.contactor.se) has left #rockbox +--- You are now known as Bagderzzz +--> wavey_ (~wavey@dlan1431.dircon.co.uk) has joined #rockbox +<-- wavey has quit (Read error: 104 (Connection reset by peer)) +<-- wavey_ (~wavey@dlan1431.dircon.co.uk) has left #rockbox +--> Zagor (~bjst@as9-5-6.k.s.bonet.se) has joined #rockbox +--> alkorr (jbcoax@srs04m-7-48.n.club-internet.fr) has joined #rockbox +<-- alkorr has quit (Client Quit) +--> wavey (~wavey@host-54.valtech.co.uk) has joined #rockbox + are you an emacs man, wavey? + yus + you also? + yep + had a problem with c-offsets-alist, but I worked it out + are you logging everything? + alan said his FAT32 code was "obsolete" but I don't understand what he meant with that +<-- wavey has quit (Read error: 104 (Connection reset by peer)) + zagor.. you still around? + yeah + heheh glad you updated the lcd.[ch] api... + now i can toss up my battery bit at some point ;) + ah, yes + i have a little something i hope to get done tonight + whats that? + umm, a surprise :) + tetris + lol.. seriously? + yeah + oh.. i was thinking... + we have a limited # of keys on the players.. we may want to assign keyboard keys to represent them for the simulator. + yes, i've been thinking that too. i was thinking some keys on the numeric keyboard + nods + that or just overload the function keys. + but the #pad would work nicely for the center control pad... + left, right, up, down. + exactly + i like that. + and 5 för the play/pause on the recorder + yup... + and even the top four for the menu keys + could even get away with - for off and + for on + right... + i don't know if numlock is a problem using as a normal key in X + i don't see why it would be... + brb..need food +<-- Zagor has quit ("bye") -- cgit v1.2.3