Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2010-12-23

00:00:15kugelor I didn't check if it's even runnable, or maybe I didn't handle all the other important bits as well :)
00:01:09jhMikeSthere's alot of important bits, like what if it exits while urgent. the core then needs updating. it needs to be irq and core safe too if it's use elsewhere. there's probably a dozen or so spots to add a tiny bit of code to handle it.
00:01:35 Quit komputes (Remote host closed the connection)
00:01:43kugelI definitely didn't add that much code :P
00:04:15jhMikeSblocked or not it shouldn't matter really. it could be only a few small spots of code. hmmm...
00:09:30 Part toffe82
00:10:39 Quit kugel (Remote host closed the connection)
00:14:05 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
00:15:56 Join kadoban [0] (
00:16:25***Saving seen data "./dancer.seen"
00:18:27wodzJdGordon|: ping
00:22:52JdGordon|wodz: semi-pong
00:25:04wodzJdGordon|: %tx is broken IMO - with line %?tx<something> it segfaults simulator with %?tx<something|> it works
00:26:10telliottAnyone use a Clip+??
00:29:38 Quit {phoenix} (Remote host closed the connection)
00:31:07 Quit TheLemonMan (Quit: free(me))
00:37:17 Quit wodz (Quit: Leaving)
00:37:25 Quit linuxguy3 (Read error: Operation timed out)
00:37:58 Quit zenlunatic (Quit: Changing server)
00:41:19 Join eWill_ [0] (
00:42:09 Join linuxguy3 [0] (
00:42:26 Quit eWill (Ping timeout: 260 seconds)
00:42:31 Nick eWill_ is now known as eWill (
00:49:35 Quit Keripo (Quit: Leaving.)
00:57:45 Quit parafin (Remote host closed the connection)
00:57:57 Join parafin [0] (
01:03:54 Join Llorean1 [0] (
01:04:21 Quit Llorean (Disconnected by services)
01:04:25 Nick Llorean1 is now known as Llorean (
01:04:27 Quit Llorean (Changing host)
01:04:28 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
01:04:55 Join eWill_ [0] (
01:05:51 Quit eWill (Ping timeout: 250 seconds)
01:06:06 Nick eWill_ is now known as eWill (
01:07:42 Part marines
01:10:01 Join JdGord [0] (
01:12:05 Quit dfkt (Quit: -= SysReset 2.53=- Sic gorgiamus allos subjectatos nunc.)
01:12:13 Quit pamaury (Remote host closed the connection)
01:17:20 Join [AndrewR] [0] (~andrewrot@
01:17:42[AndrewR]hi, does anyone know how long a tick is in void (*sleep)(int ticks); ?
01:18:40JdGordArg... 1/100th of a second
01:19:23[AndrewR]I used my workout app for the first time at the gym last week :D
01:19:33[AndrewR]it was useful
01:19:52[AndrewR]told me what weight to use, how long to rest between sets
01:20:09preglowthere's a workout app? :>
01:20:15[AndrewR]still some bugs to work out then I think I'll make a youtube vid to show, see if anyone else is interested
01:20:19[AndrewR]I made one
01:20:26preglowthis is interest
01:21:07[AndrewR]1 sec
01:21:19[AndrewR]I'm currently on a train with really crappy internet
01:21:39[AndrewR]my app is a combination of a ruby on rails app and rockbox plugin
01:22:04preglowi assume the ruby part isn't rockbox hosted, heh
01:22:26[AndrewR]no, people have to host that themselves
01:22:49[AndrewR]it's really more a proof of concept / something people with some development skills can set up themselves
01:23:53[AndrewR]I'm not a C programmer
01:23:53[AndrewR]so it's a bit messy
01:24:04[AndrewR]but it's working on my sansa
01:24:38preglowsounds like a good concept anyway
01:24:45preglowdoes the rest between sets matter much anyway? :>
01:24:54[AndrewR]but I have actually used it in the gym
01:25:10[AndrewR]no, not the specific time
01:25:11[AndrewR]but I don't want to think when I'm working out
01:25:11[AndrewR]it beeps at me to tell me to go again
01:25:18[AndrewR]I don't like staring at the clock
01:25:29preglowheh, fair point
01:25:32[AndrewR]basically I set up the workouts in the rails app
01:25:49[AndrewR]then the plugin is a workout "player" that has a list of the exercises and sets in your workout
01:25:54preglowyeah, makes sense
01:26:06[AndrewR]and advances through stages: setup, workout, rest.. repeat
01:26:07preglowassuming you're the only person in the gym :>
01:26:11[AndrewR]pretty simple
01:26:17[AndrewR]oh, there's lot of ppl there
01:26:21 Quit JdGordon| (Quit: Lost terminal)
01:26:23[AndrewR]it just looks like I'm messing with my playlist
01:26:38[AndrewR]I use earphones
01:26:40preglowjust thinking as long as nothing's taken
01:26:50preglowthere could be a pause button anyway
01:27:01preglowi don't care if i look stupid while working out, so that's not a problem :P
01:27:02[AndrewR]yeah, it's not really flexible for subbing in different exercises on the go
01:27:13[AndrewR]you can skip to a different spot in the workout
01:27:34[AndrewR]I also store the workout data on the pandora, and then sync it with the rails app after
01:27:54[AndrewR]I want to graph my progress
01:28:22preglowyeah, sounds truly sweet
01:28:23[AndrewR]I currently have it at a point where I could do that, if I had more data
01:28:29preglowcan you update it if you modify your program on the go?
01:28:44[AndrewR]I'm not sure what you mean
01:29:05[AndrewR]I run a capistrano task to sync the player with my rails db
01:29:18preglowwell, deviate from my planned setup, up the weights, run longer
01:29:21[AndrewR]it uploads the logs from the player, then downloads any new workout data (exercises, etc)
01:29:26preglowif i decide to do that, i want to have it logged
01:29:31 Join JdGordon| [0] (
01:29:31 Quit JdGordon| (Changing host)
01:29:31 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
01:29:32[AndrewR]ah. no, not there yet :)
01:29:43[AndrewR]you're right, I want that feature too
01:29:51[AndrewR]but I had to trim features to get something working
01:29:56[AndrewR]so right now it's a strict player
01:29:59preglowyeah, but it should be doable
01:30:07preglowand it would be quite killer if possible
01:30:12[AndrewR]if you change it, you have to remember and modify it in the rails app
01:30:12preglownot much else i want to be able to do :)
01:30:13[AndrewR]yeah, it's do-able
01:30:38[AndrewR]it's a bit tricky for me to know how that should work though, with respect to goals
01:30:55[AndrewR]if you change from 50lb to 70lb at the gym, does that change your goal or just log it?
01:31:06[AndrewR]either is doable, just not sure yet how that should work
01:31:15[AndrewR]let me make a backup of my rails db
01:31:19[AndrewR]then I'll show you that
01:31:20preglowit should just log it, really
01:31:26[AndrewR]yeah, probably
01:31:41preglowwhether i want to change my goal or not would vary a bit
01:31:57[AndrewR]the goals stuff right now are simple linear functions
01:32:05[AndrewR]each workout has an integer number
01:32:06preglowand would not be something i want the app to decide without further input anyway
01:32:07[AndrewR]starting at 0
01:32:13saratogapreglow: you wrote the EQ code right?
01:32:17preglowsaratoga: correct
01:32:25preglowsaratoga: at least the dsp part of it, safetydan did the ui part
01:32:26[AndrewR]then you say "weight", 50 + 2*n, kinda thing
01:32:34saratogai've been meaning to go through and update the ASM for our DSP code for arm9/11
01:32:41saratogai had a few questions
01:33:04preglowi remember most of it, so you're in luck
01:33:11saratogai noticed the replaygain scale values are 32 bit, that could probably be 16 bit, but what about the EQ filter coefficients?
01:33:28saratoga16 bit is much faster on arm9e and arm11
01:33:37preglowthey could be 16 bit, but not with any ease
01:33:48preglowpeaking filters use quite a few bits of integer precision
01:33:53preglowbut the frac bits are in many ways what matters
01:34:28preglowand with "peaking" i mean "shelving"
01:34:39[AndrewR]preglow my workout rails app is at please don't modify stuff, it's not passworded :S
01:34:54saratogai'm not really familiar with the EQ, whats the difference between those?
01:35:10preglowsaratoga: peaking has 0db gin at both 0 hz and max hz
01:35:17preglowand whatever you want in between
01:35:18[AndrewR]currently it's simple rails scaffolding, I hope to make it a nicer interface
01:35:29preglowshelving filters have x db gain at either 0 hz or nyquist
01:35:32preglowand 0 at the other end
01:35:45saratogathese are IIR filters?
01:36:39preglowyeah, most eqs are
01:36:51saratogaso rounding errors will compound
01:36:59preglowin a way
01:37:16preglowrounding errors in the coefs will only give a different frequency response
01:37:22preglowwith wrong freqs and gains, that kind
01:37:44preglowrounding errors in the filter calculation will manifest as noise and, with bad luck, as limit cycles
01:37:49preglowwhich are bad
01:38:01preglowyou shouldn't quantize filter coefs to 16 bits unless you really know what you're doing
01:38:04preglow16 bit audio data is ok
01:38:17preglow[AndrewR]: i'll be good :)
01:38:23saratogaarmv5e has a 32x16 multiply accumulate with 32 bit accumulation
01:38:40saratogaso if possible i'd like to use them
01:39:11saratogai guess if theres no obvious way to use faster multiplies, i could just go through and reschedule for arm9's longer pipeline
01:39:19saratogathe current asm is very arm7ish
01:39:28preglowthat's all i thought of when coding it
01:39:41preglowyou can pipeline for arm9 without impacting arm7 very foten
01:39:42preglowoften too
01:41:51saratogapretty much always
01:42:12saratogai dont' think theres anything that benefits arm7 while hurting arm9/9e/11
01:42:15preglowbut yeah, using the new dsp instructions would be very good, they're fast as hell
01:42:24preglowi did think about it, and it won't be trivial
01:42:34preglowquantizing the coefs to 16 bits would be the hard way
01:42:53saratogai think at 16 bits it'd take about 1MHz per EQ band on the Clip+
01:43:44saratogabut i guess even without it the eq is so fast its not a big deal
01:44:29preglowfaster is better ;)
01:44:34preglowi'd def go for 16 bit audio first
01:44:42preglowthat's what we mostly use anyway
01:45:30 Join ampdj89 [0] (~merrychri@
01:45:44saratogayeah that might work
01:45:54 Nick ampdj89 is now known as Guest20103 (~merrychri@
01:46:04saratogahow far into the DSP engine is the EQ?
01:46:17preglowi really don't remember, quite late i think
01:46:23preglowshould be easy to find out from dsp.c
01:47:34Guest20103i know this has been answered but ipod classic is nearing its life as speculated, wouldnt it be effective to port rockbox to the 6th classic ipods? the iOS has much complicated firmware lock yet there lots of support.
01:47:45preglowGuest20103: effective, yes, easy, no
01:47:51jhMikeSlet's just go 8 bits, it should be enough for everyone, the rest is just snake oil *runs and hides*
01:47:58Guest20103hm, wat makes it different
01:48:24preglowsaratoga: but yeah, i would definitely recommend going for a 16 bit audio version, you would have no need to think about the filter coefs at all (the tricky part), and the perf gains should be equivalent
01:48:46preglowGuest20103: i don't believe we have any good way of running code on the 6th gen classic
01:48:53preglowi could be wrong here
01:49:11saratogatheres been a way to run code on it for a while, just until now no interest on working on it
01:49:26saratogaTheSeven was working on it, not sure if he still is, hes got like 8 ports going at once :)
01:49:44JdGordon|he's cheating.. all the ports are realy the same hardware
01:50:02saratogawell vaguely similar anyway
01:50:19saratogapreglow: ok i'll probably look at that after i finish libmad
01:50:32Guest20103so is rockbox and extremely mofified linux port? if it would be atleast have 60% linux... then it would have been ported right, or is it a kernel problem
01:50:39saratogawe're not linux
01:50:49saratogano idea why people thing we are
01:50:49Guest20103oh ok then
01:51:10preglowsaratoga: at least that way you would discover if the gains are worth it without all the fuss with mucking about with filter coefs
01:51:10scorchesaratoga: because only insane people would try to write their own OS
01:51:13preglowwhich can be very messy
01:52:03saratogawell i'm sure i'll get roughly a 2x speed up
01:52:14saratogaif its worth it depends on how much you care about making something fast even faster
01:52:29preglownot so sure about 2x, but worth a shot
01:52:37preglowless than 2x would still be wonderful
01:53:32preglowthe 32x16 instructions can function in saturating mode, rite?
01:54:19saratogai don't think so
01:54:27preglowah, that might be the 16x16 one
01:54:39jhMikeSI don't think any multiplies do, just adds and subtracts (if it starts with "q")
01:54:42saratogausually its 2x since each multiplication takes half as long, and you can load two coefficients per ldr
01:55:00preglowa fair point
01:55:07saratogaalthough unless we can fold truncation into some earlier step that'll slow things down
01:55:14saratogaand i guess if people want dithering that should be done first
01:55:19preglowyeah, you're right
01:55:21saratogacould be a pain to make this happen actually
01:55:24preglowfilter coefs don't need loading
01:55:40saratogaoh because theres so few?
01:55:51saratogabut i guess thats one more reason to do the audio samples
01:56:13preglownot really, if you can fit 2x the coefs into the same space, you _could_ do stereo processing
01:56:16preglowthe savings would be immense
01:56:17saratoga16 bit math is really nice since it basically doubles the number of registers, allowing you to pretend arm is a sane isa
01:56:19 Quit [AndrewR] (Quit: [BX] Did somebody say BitchX?)
01:57:06saratogai always wondered why we were the only people on earth writing stuff in arm asm, then i started doing it and i'm not surprised
01:57:16saratogaor rather no longer surprised
01:57:22preglowdsp and asm goes hand in hand
01:57:27pregloweven on x86
01:58:30preglowanyway, one of my main criteria when writing those asm loops was keeping the coefs in regs at all times
01:58:37preglowso i only have to load samples
01:58:44saratogayeah makes sense
01:58:50preglowit just about works because they are so few
01:58:55jhMikeSpreglow: what were you planning for the full-duplex, if you don't mind me asking?
01:58:57preglowbut i had to cut corners in the arm case
01:59:17preglowjhMikeS: nothing major, really, i just bought myself a mixer and wanted to get my fx plug working again
01:59:33preglowbut i can't make it work after the api changes
01:59:57preglowprobably just me doing stupid shit
02:00:51jhMikeSCan't do processing in the isr anymore but other than that, it should be ok. since we have pitch detector, doing playback too should be all the same
02:01:16preglowi'm not doing it in the isr, i'm basically still using your old framework plugin, which uses thread and circular buffers
02:02:10preglowthe pitch plugin does much the same thing, but with two buffers instead of 2^x
02:02:16preglowand no playback, ofc
02:02:56jhMikeSyeah, no playback but playback would be the same as any other plugin if it did, so hmmm....
02:03:29preglowlike i said, probably just me doing stupid shit, i just lost all my will and/or ability to debug/hack lowlevel stuff to figure out how and why it fails
02:03:38jhMikeSI made pd do that. at first it was full starting and stopping recording on every frame
02:04:22preglowi did look over the last sound api changes and agreed with myself that not enough changed for full duplex to have broken
02:04:46 Join eWill_ [0] (
02:04:53 Quit Kitar|st (Ping timeout: 260 seconds)
02:05:13preglowsaratoga: any targets in particular that would benefit from arm5 opts, btw?
02:05:26preglowapart from all the new ones, that is, but they have mhz to burn anyway :P
02:05:35 Quit eWill (Ping timeout: 240 seconds)
02:05:43 Nick eWill_ is now known as eWill (
02:05:50saratogapreglow: Clipv2/Fuzev2/Clip+/GigabeatS/D2/Android
02:05:56saratogaprobably some up and coming targets as well
02:06:03*jhMikeS wasn't aware that there were armv5 with the same dsp instructions as imx31, interesting
02:06:09saratogai don't think anyone is still making armv4 players
02:06:33saratogajhMikeS: ARMv5E has most of the DSP instructions as ARMv6 but not all
02:06:56saratogathe arm docs are actually pretty confused with some saying 5E has the v6 ops, but they don't work on my player
02:07:38saratogaso i can't do the top half 32 bit multiply for instance on the clip+
02:07:38preglowarmv6 does the simd 16x16 stuff, no?
02:07:56saratogayeah it also adds packed 16 bit multiplies, but they're pretty useless for audio IIRC
02:08:02saratogasince i don't think they saturate
02:09:21preglowloads of the 32x16 ones seem to update a saturate flag
02:09:29preglowbut i dunno how much that helps
02:09:40saratogaactually SMLALD might be useful
02:09:41 Join Kitar|st [0] (
02:09:54saratogafor complex multiplies, but its only 16 bit so rounding might be ugly
02:10:25saratogaheh its a two cycle instruction on arm11
02:10:42saratogaso not very exciting
02:11:23preglowhow i wish people would copy emac
02:11:32saratogaso unless i'm missing something its not even faster then just doing two multiplies
02:11:40saratoga(which would be more accurate)
02:11:51preglowit probably is taking into account loads
02:12:02saratogayeah theres that
02:12:22preglowmemory pretty much always ends up being the limiting factor
02:12:24saratogaIIRC the only reason we use the armv4 motion comp on armv5 is that the armv6 code uses those packed multiplies
02:12:40saratogamaybe i should look and see if the armv6 code is actually faster then just using armv5 ops
02:12:51jhMikeSI think the packed multiplies add to 32 bit accums, iirc
02:13:32 Quit simonrvn (Read error: No route to host)
02:14:28 Quit YPSY (Ping timeout: 245 seconds)
02:14:50saratogayeah but one of the operands can be 32 bit, so its still less accurate
02:15:55saratogaoh wait
02:15:58jhMikeSthere's stuff like smuad, and then smulwb(t) which is 32X16->48, I think you mean
02:16:07saratogatheres a 16 bit multiply with 64 bit accumulate
02:16:29***Saving seen data "./dancer.seen"
02:17:00saratogasmulwb/smlawb is the main one i've been using for atrac and mp3, its really just a 32 bit accumulate with the top 32 bits of the 48 bit product
02:17:02jhMikeSheh, is there? I don't think ever I needed it for what I did
02:17:18saratoga"Signed Multiply-Accumulate with 16-bit operands and a 64-bit accumulator. "
02:17:25saratogabeats me what you'd want to use that for
02:17:37preglowsaratoga: a very clever scheme, the top bits are what you'd use anyway
02:17:48preglowand basically what we do on emac
02:17:53saratogayeah smlawb is amazing useful
02:18:03saratogai can't believe it took until armv6 to get a 32 bit version of it
02:19:24saratogai kind of get the feeling there was one DSP guy at arm ltd before 2002 or so, and everyone there didn't like him because he was creepy
02:19:34preglowif there's one thing i learnt looking at emerging cpu instruction sets, it has to be the fact that instruction set designers never understand what you actually need :P
02:19:37preglowjust look at sse
02:19:44preglowand all revisions up and until the very recent ones
02:19:53saratogai actually don't know anything but arm
02:20:20preglowi'm not an expert, but i always did pc coding, and watched every sse update with great anticipation :P
02:20:31preglowfollowed by violent bouts of cursing
02:20:37 Join simonrvn [0] (
02:20:42saratogai think one of the revisions of sse4 has a multiply accumulate instruction
02:20:50jhMikeSpreglow: I thought they did extensive testing to find the most "useful" stuff. :)
02:20:55preglowthey got the point in the end
02:21:00preglowamd tho
02:21:07preglowgot it in the first version of 3dnow
02:21:09 Join Ypsy [0] (
02:21:20 Nick Ypsy is now known as YPSY (
02:21:21preglowno wonder those guys were the ones who invented amd64
02:21:25saratogawhatever their post SSE ISA is called is supposed to have 256 packed multiply accumulates, so i guess that doesn't suck
02:21:32saratoga256 nbit
02:21:50saratoga(sorry just got contacts and am dizzy)
02:22:23 Quit JesusFreak316 (Ping timeout: 245 seconds)
02:24:00*jhMikeS just wants to find out why the SPC codec is dead on arm6 when it worked before even after asm upgrades :\
02:25:13saratogagcc updates?
02:25:46preglowlikely suspect
02:26:37jhMikeSnot sure. I'm not sure if somehow I missed the problem with going eabi or not. rockbox just before eabi won't play anything whatsoever it seems.
02:31:47saratogadoes the gigabeat S have frequency scaling yet?
02:33:13jhMikeSyes, it's automatic though
02:33:42jhMikeSand voltage. voltage seems to make all the difference for any gain, frequency not much at all
02:34:02saratogayeah frequency doesn't make much difference on AMSv2 either
02:34:21saratogais there any way to force a lower max clock? it'd be neat to test arm11 at low clock speeds
02:34:36saratogaalso, do you remember what the memory clock is?
02:35:03jhMikeSthat's tricky because there's a bug that probibits lowing the core relative to the bus clocks so much
02:35:57preglowi gotta get some sleep done, have a nice night/day guys :)
02:35:58jhMikeSyou'd have to reclock it all. right now it just uses divider scaling. pll relock has other issues with the audio
02:36:05jhMikeSpreglow: gnight
02:36:18saratogadoes it ever run with a divider of 1?
02:36:59jhMikeSsaratoga: at high loads. that's overdrive (528MHz) which also needs elevated voltage, which also degrades the chip faster :\
02:37:29saratogai mean with the cpu clock and memory clock equal
02:37:56saratogaideally i'd like to have test codec run with the memory clock sync and with the CPU clock maxed
02:37:56jhMikeSno, the errata apparently prohibit that ratio
02:38:09saratogawhats the smallest one?
02:38:29jhMikeS2:1 for 132MHz core, 66Mhz memory
02:38:40 Join antil33t [0] (
02:39:50jhMikeSit's possible to turn off frequency scaling then set the working point manually. the apis exist in the driver but they're not used atm
02:42:39 Quit saratoga (Quit: Page closed)
02:48:49 Join saratoga [0] (62e16704@gateway/web/freenode/ip.
02:52:25 Join designate72 [0] (
03:02:50 Join madalu [0] (~user@unaffiliated/madalu)
03:08:35 Quit DerPapst (Quit: Leaving.)
03:12:49 Join [Saint] [0] (S_a_i_n_t@
03:17:40 Quit saratoga (Ping timeout: 265 seconds)
03:17:44 Quit mortalscan (Ping timeout: 265 seconds)
03:19:05 Join mortalscan [0] (~mortalsca@
03:20:45 Join [AndrewR] [0] (
03:23:11 Quit JdGord (Quit: Bye)
03:26:18 Join Judas_PhD [0] (
03:32:20 Join CaptainKewl [0] (
03:33:54CIA-7New commit by rmenes (r28879): Disktidy plugin: add .Spotlight-V100 directories ...
03:35:50CIA-7r28879 build result: All green
03:37:21 Join saratoga [0] (62e16704@gateway/web/freenode/ip.
03:41:54 Join mordocai [0] (~mordocai@
03:50:14 Join The_Pwny [0] (
03:52:35 Quit telliott (Ping timeout: 240 seconds)
03:54:16 Join fyrestorm [0] (
03:54:28jhMikeSis there a define saying if the eabi compiler is being used or not?
03:58:17JdGordon|prob best to ask in #gcc
03:58:28JdGordon|isnt there a #define for the gcc version string?
03:59:45jhMikeSfor that, yes. is the number a reliable way though
03:59:56 Quit madalu (Read error: Connection reset by peer)
04:00:15JdGordon|well considering we specify the compiler to use it is probably safe
04:00:17JdGordon|or good enough
04:00:37JdGordon|just #error out on unknown compiler os it isnt forgotten about
04:00:45jhMikeSI try non-eabi for gigs and the ldmpc macros don't work, nor would the init section
04:04:00 Quit Judas_PhD (Remote host closed the connection)
04:04:07 Quit TheSeven (Disconnected by services)
04:04:08 Join The_Seven [0] (~TheSeven@rockbox/developer/TheSeven)
04:08:21 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
04:11:28[Saint]What are peoples views about settings/subdirs with only one option...and what to do about them.
04:11:42[Saint]To me, it seems like it could be avoided with some thought.
04:13:25[Saint]My view is that if only one choice is presented to the user, it really isn't a choice at all and just adds noise to the menu. There isn't *too* many places in the menu where this happens, but there is a few.
04:14:25JdGordon|are you sure you want to go down that particular rabbit whole?
04:15:51[Saint]I'm sure it's not *that* much a big deal, makes perfect sense to eliminate cases where there isn't 2 or more choices available.
04:15:56JdGordon|[Saint]: just remember that while a submenu might only one have option on your device, it might have 3 on another
04:16:31[Saint]JdGordon|: I'm aware of that.
04:16:33***Saving seen data "./dancer.seen"
04:16:39[Saint]But, surely that can be compensated for.
04:16:58JdGordon|yes, but it gets complicated to do it more than a few times
04:17:06 Quit designate72 (Ping timeout: 255 seconds)
04:17:07JdGordon|it is already done for the time&date screen
04:17:32[Saint]I think it would be worthwhile, once it was done it would be easy to maintain.
04:17:52[Saint]it's just the getting it done initially thing...
04:19:22 Join lwhalen [0] (
04:19:32lwhalenhey all
04:19:59[Saint]Welcome, if you have a question/query about Rockbox..just ask away.
04:20:12lwhalenI am an unfortunate owner of one of those 120GB iPod Classics, and after some judicious Googling I've found that it's going to be pretty much impossible for RockBox to ever be ported to it
04:20:30[Saint]That is quite incorrect.
04:20:49lwhalenthere seems to be no ipod 6G support, per the website...
04:21:00[Saint]While there is no ROckbox port for it, that is not because it is "impossible".
04:21:13[Saint]It simply is a huge amount of work, and isn't done yet.
04:21:23lwhalenmore judicious Googling seems to say that it's because the ROM image is encrypted
04:21:40lwhalenso, 'mathematically improbable' might've been the better term :-)
04:22:12[Saint]The Nano2G also has encryption...and a Rockbox port.
04:22:16lwhalenI was wondering, what would you guys recommend for an ipod alternative that holds at least 120GB, and I can hook up to my car stereo's aux input?
04:22:21lwhalenheh, fair enough :-)
04:22:42[Saint]Until not so long ago, there wasn't a known exploit for the Classic but there has been one for some time now.
04:23:17[Saint]The_Seven has been working on the Classic for ~3 weeks...perhaps longer.
04:23:20lwhalenis there a roadmap/feature request/etc for Classic support?
04:23:24*lwhalen does a happy dance
04:23:28[Saint]And is running arbitrary code for some time.
04:23:48lwhalenwell alright then, slap mah 'fro and call me Susan
04:24:17[Saint]To follow the progress of the Classic it would be best to follow the freemyipod wiki/svn
04:24:58[Saint].org no?
04:25:03saratogai don't remember
04:25:12saratogahonestly just punch it into google
04:25:56[Saint]emBIOS is really the relevant freemyipod work in this case, though.
04:26:22[Saint]Not sure how up to date the pages are about the Classic work though.
04:26:23lwhalenhaha, epic AND awesome
04:26:29lwhalenI think I owe somebody some beer
04:26:48[Saint]My address is on my wiki page ;) seems to think that the Classic 2G (which I believe is what I own - 120GB) is coming along nicely
04:27:17lwhalensaint - link?
04:27:35[Saint]I was kinda kidding...
04:28:13*lwhalen shrugs
04:28:22lwhaleny'all should have Amazon/B&N wishlists up ;-)
04:28:44saratogaif you want to donate, send someone on that project an ipod
04:28:46[Saint]Rockbox has a donate button on the mainpage.
04:29:18JdGordon|lwhalen: you can buy me a archos 101 if you want? :D
04:29:33lwhalenthis brings me back to my original question...
04:29:55lwhalenwhat would you guys recommend for an ipod alternative that holds at least 120GB, and I can hook up to my car stereo's aux input? about the same size/portability as an ipod classic would be ideal
04:30:01[Saint]but, indeed...that work just basically wouldnl't exist if some kind soul didn't send a slightly messed up Classic to The_Seven to work on.
04:30:21 Join simonrvn_ [0] (
04:30:38[Saint]lwhalen: An iPod Video?
04:30:44saratogalwhalen: this isn't really the right place to ask about mp3 players, try rockbox-community
04:30:49[Saint]Or..did you specifically not want an iPod?
04:30:53 Quit Barahir (Read error: Operation timed out)
04:31:03lwhalensaratoga - my apologies, I take it this is more of a dev-centric channel?
04:31:13lwhalensaint: I was looking for specifically not-Apple
04:31:22lwhalenand not-Microsoft either ;)
04:32:09 Quit simonrvn (Ping timeout: 240 seconds)
04:32:09 Nick simonrvn_ is now known as simonrvn (
04:34:46 Join Barahir [0] (
04:40:30 Quit amiconn (Disconnected by services)
04:40:30 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:40:38 Quit pixelma (Disconnected by services)
04:40:40 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:40:42 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:40:47 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:46:58 Part lwhalen
04:54:53 Quit GeekShadow (Quit: The cake is a lie !)
04:54:58 Join nidzimjelp [0] (
05:01:00[Saint]Welcome, if you have a question/query about Rockbox..just away.
05:01:43jhMikeSwtf is rockbox and why, why the hell should I give a crap?
05:02:52[Saint]haha..oops, nice typo. "just ask" rather.
05:05:18nidzimjelpAre "viewports" also used to themes to set the coordinates for rockbox to start displaying main menu text?
05:08:02nidzimjelpThank you.
05:14:35jhMikeSargh, you can't even compile this thing with −−no-eabi anymore :\
05:17:21[Saint]nidzimjelp: For that, you need to set a UI Viewport, specifically.
05:29:28 Quit Rob2222 (Ping timeout: 276 seconds)
05:30:30 Quit Horscht (Quit: Verlassend)
05:32:11 Quit advcomp2019 (Read error: Connection reset by peer)
05:32:58 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
05:38:13 Quit BHSPitMonkey (Ping timeout: 240 seconds)
05:39:14 Quit Pun (Ping timeout: 250 seconds)
05:40:01 Quit [AndrewR] (Quit: BitchX: it's all day strong, all day long)
05:40:46[Saint]I'd like to include a small addition to CabbieV2 if noone is going to cry about it.
05:41:14[Saint]I'd like the "time remaining" under the progressbar to have a -
05:41:19JdGordon|HAHAHAHA what ever gave you that idea?
05:41:46[Saint]what, the "-", or noone crying about it? ;)
05:41:52JdGordon|the latter
05:42:19[Saint]I was thinking of including it when I fix up the volume icon to change to text whilst it's being changed.
05:42:26 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
05:42:28jhMikeSwhat, without the "-" I might be confused?
05:43:00[Saint]I think it looks more complete with it, personally.
05:43:36[Saint]makes it more obvious that it is a negative value, and counting down.
05:43:49LloreanIt's not a negative value.
05:44:00LloreanIt's 4 minutes remaining, not negative 4 minutes remaining.
05:44:33jhMikeST -4:00 until launch
05:44:48[Saint]haha, I was just thinking of that example.
05:45:44LloreanYeah, but that's a countdown rather than a duration remaining. You have X minutes to Event, rather than X minutes left of Song.
05:45:51jhMikeSyou could think of it as -4:00 into the next song, then it makes sense
05:46:24[Saint]jhMikeS: That is how I look at it really, but like all things I thought I'd voice it first before wasting time with a patch.
05:46:27LloreanYeah, my point though is that it's currently presented as time remaining.
05:46:45nidzimjelpAnother question if I may. If the font is specified in the .cfg file, how should the "ui viewport" character for font be? 1 or "-"?
05:46:56LloreanI'd like to see a fix that shows numeric volume while it's being changed though.
05:47:00JdGordon|both would work nidzimjelp
05:47:07nidzimjelpAssuming 0 is for some default font used by RB.
05:47:09[Saint]nidzimjelp: both.
05:47:16[Saint]1 == user font.
05:47:22LloreanAs to the time remaining, I'm mostly indifferent, but it seems like something that's more worth discussing since it's very subjective how you interpret it and it's not something worth changing arbitrarily without getting opinions - nothing's broken as it stands.
05:47:32JdGordon|[Saint]: not both, that obviously cant work... either would though
05:48:24[Saint]well, yeah...worded poorly. either, was what I meant. I guess it wasn't as obvious to others what I meant as it was myself.
05:57:34 Part Guest20103 ("Leaving")
06:03:28 Join JesusFreak316 [0] (
06:06:00 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
06:06:05nidzimjelpOk, something is weird. "ui viewport is written right but whenever I select the theme or switch the base skin to this theme the menu items disregard the view port. Would you have any ideas as to what it may be?
06:06:50[Saint]Not without seeing the code, no.
06:06:55[Saint]Pastebin your theme.
06:09:34nidzimjelpThis is what it should look like:
06:10:51[Saint]your UI viewport statement is quite incorrect.
06:11:01nidzimjelpBut the "Rockbox, Files, Database, etc" move up and appear behind the ... oh
06:12:01[Saint]it should look as: "%Vi(-,X,X,X,X,X)%Vf(XXXXXX)%Vb(XXXXXX)"
06:12:29[Saint]using "-" as the identifier for the UI viewport makes sure that it is used by default.
06:13:52nidzimjelpOk, give me a sec.
06:15:07[Saint]ui viewport: %Vi(-,5,50,166,150,-)%Vf(333333)%Vb(-) should work fine.
06:16:37***Saving seen data "./dancer.seen"
06:20:15 Quit JesusFreak316 (Ping timeout: 245 seconds)
06:23:05 Join JesusFreak316 [0] (
06:30:31jhMikeShmmm...gcc seemed to overwrite the contents of an input operand into an output that was assigned the same register, but before the input was referenced :\
06:33:10 Quit xavieran (Ping timeout: 276 seconds)
06:34:21 Join xavieran [0] (
06:35:32nidzimjelpFinally got it working.
06:35:41nidzimjelpThank you.
06:35:48[Saint]No problem.
06:36:16nidzimjelp Though it wasn't in the cfg file
06:36:47nidzimjelpit was missing that statement from the sbs
06:36:52[Saint]well, the line for UI Viewport in your .cfg was certainly incorrect.
06:37:46[Saint]Oh, wait...I think I confused myself. Yes, it should have looked as I stated in the .sbs
06:37:51nidzimjelpThe thing is that after I tried it and it didn't work, I checked another theme that has about the same style and the uivp appeared the same
06:38:01nidzimjelpno problem
06:38:06nidzimjelp thank you
06:38:30nidzimjelpIf you hadn't mentioned that I wuold never have been able to check for that line
06:39:15[Saint]I opened an .sbs instead of a .cfg file when I was checking how it should look. No idea how I mixed those two up, sorry for that.
06:39:30[Saint]Glad you got it working, though.
06:40:28nidzimjelpDo you know anyone in charge of the themes on the site
06:41:04JdGordon|whats the problem?
06:41:17nidzimjelpThis theme:
06:41:31[Saint]what about it?
06:41:56nidzimjelpThe sbs file needs this line "%Vi(-,5,50,166,150,-)%Vf(333333)%Vb(-)" added.
06:42:23[Saint]it's up to the original author to update it.
06:42:31JdGordon|cant anyone upload updateS?
06:42:46[Saint]not to the same theme, anyway.
06:43:02nidzimjelpI'll email him asking him to update it then.
06:43:02[Saint]nothing stopping you from uploading it under another name though.
06:43:14[Saint]probably the best idea.
06:43:36nidzimjelpOk, well thank you.
06:43:49JdGordon|maybe we should setup something to allow minor mods
06:43:53nidzimjelpI'll be going now, you've been great help. bye
06:44:07[Saint]JdGordon|: Not a good idea IMO
06:44:29 Quit nidzimjelp (Quit: CGI:IRC (EOF))
06:48:34 Quit factor (Ping timeout: 250 seconds)
06:49:34 Quit Strife89 (Quit: Logging off.)
06:50:18 Join utanapischti [0] (~username@
06:53:19 Quit sasquatch (Ping timeout: 265 seconds)
07:08:28 Join factor [0] (
07:13:26 Join einhirn [0] (
07:23:57 Quit JdGordon| (Quit: leaving)
07:27:28 Join anewuser [0] (anewuser@unaffiliated/anewuser)
07:31:09 Join JdGord [0] (~jonno@
07:36:32 Quit BHSPitMonkey (Read error: Connection reset by peer)
07:50:17 Join n1s [0] (~n1s@rockbox/developer/n1s)
07:52:12 Quit JdGord (Quit: Bye)
07:54:06 Quit jhMikeS ()
07:54:53 Join jhMikeS [0] (
07:54:53 Quit jhMikeS (Changing host)
07:54:53 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
07:59:50 Quit timccc (Ping timeout: 240 seconds)
08:00:03 Join LinusN [0] (~linus@rockbox/developer/LinusN)
08:02:10 Quit factor (Ping timeout: 240 seconds)
08:16:38***Saving seen data "./dancer.seen"
08:16:41 Join ender` [0] (
08:20:54 Join JdGordon [0] (3a6c66cc@gateway/web/freenode/ip.
08:21:55 Quit JesusFreak316 (Ping timeout: 245 seconds)
08:23:44 Join timccc [0] (~timccc@
08:33:02 Quit CaptainKewl (Read error: Operation timed out)
08:36:24CIA-7New commit by jethead71 (r28880): SPC Codec: Was broken on ARMv6 since EABI switch and somehow I failed to notice in the first place (and so did eveyone else it seems :). Some ...
08:38:31CIA-7r28880 build result: All green
08:52:00 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:10:43 Join Bagder [0] (
09:12:29 Join factor [0] (
09:19:47 Join petur [0] (d408b802@rockbox/developer/petur)
09:20:32 Quit simonrvn (Quit: ZNC -
09:23:02 Quit bertrik (Quit: :tiuQ)
09:44:30 Quit factor (Ping timeout: 264 seconds)
09:49:07 Quit JdGordon (Ping timeout: 265 seconds)
09:57:55 Join timccc1 [0] (~timccc@
09:58:17 Quit timccc (Read error: No route to host)
10:09:00 Quit xavieran (Ping timeout: 272 seconds)
10:12:23 Join xavieran [0] (
10:16:39***Saving seen data "./dancer.seen"
10:21:36 Quit xavieran (Ping timeout: 240 seconds)
10:22:28 Join xavieran [0] (
10:22:39 Quit ranmachan (Ping timeout: 250 seconds)
10:23:53 Part LinusN
10:24:49 Join factor [0] (
10:25:02 Quit kadoban (Ping timeout: 265 seconds)
10:33:47 Quit n1s (Quit: Lämnar)
10:37:45 Join wodz [0] (
10:37:51 Quit factor (Read error: Connection reset by peer)
10:37:54wodzZagor, Badger: ping
10:41:28 Join pamaury [0] (
10:41:28 Quit pamaury (Changing host)
10:41:28 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:55:18 Join factor [0] (
10:57:28 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
10:59:29 Join TheLemonMan [0] (
11:05:54 Quit simabeis_ (Ping timeout: 240 seconds)
11:05:59 Join simabeis [0] (
11:09:04 Join {phoenix} [0] (
11:12:29 Join stoffel [0] (
11:20:37 Join mystica555_ [0] (
11:24:47 Join {-phoenix-} [0] (
11:26:30 Quit MagusG (Read error: No route to host)
11:26:42 Join MagusG [0] (
11:27:28 Quit {-phoenix-} (Remote host closed the connection)
11:27:49 Quit kevku (Remote host closed the connection)
11:28:13 Quit {phoenix} (Ping timeout: 255 seconds)
11:30:43Stummiis the tracker the right place to moot a change on makefile?
11:32:09StummiI made some changes on root.make and, what I want to put up for discussion anywhere before eventually commiting it
11:32:31 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:39:41 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:46:20 Quit xavieran (Ping timeout: 260 seconds)
11:46:48 Join n1s [0] (~n1s@rockbox/developer/n1s)
11:47:19 Join xavieran [0] (
11:47:57 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
11:51:49 Join kugel [0] (~kugel@
11:51:50 Quit kugel (Changing host)
11:51:50 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:51:56 Quit xavieran (Ping timeout: 245 seconds)
11:52:58 Join xavieran [0] (
11:54:36 Quit n1s (Ping timeout: 240 seconds)
11:56:10 Join casainho [0] (
12:03:46linuxstbStummi: Yes, you should post the patch there. But you may also want to try to get people to talk about it here or on the dev mailing list.
12:05:48 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
12:07:05 Quit dfkt (Read error: Connection reset by peer)
12:13:11 Join JdGord [0] (~jonno@
12:13:46Stummiok, patch added :)
12:14:27Stummibtw, a bot, what would post actions on tracker here (like CIA with the CVS does) would be a nice thing
12:16:40***Saving seen data "./dancer.seen"
12:17:02JdGordwodz: sorry about earlier
12:17:04 Quit kugel (Read error: Connection reset by peer)
12:17:13JdGordWhat's %tx. ?
12:17:30JdGordAnd yes if it is segfaulting it needs fixing
12:17:49JdGordFile a bug. I might be able to fix it on the plane tomorow
12:18:49gevaertsStummi: yuou know there's a mailing list for flyspray updates?
12:19:05 Quit wodz (Quit: Leaving)
12:23:34Stummigevaerts, yes
12:25:59 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:26:42 Part LinusN
12:27:12 Quit dfkt_ (Read error: Connection reset by peer)
12:27:42 Quit mystica555_ (Ping timeout: 272 seconds)
12:29:15 Quit JdGord (Read error: Connection reset by peer)
12:30:09*gevaerts would call the thig "symlinkinstall". "link" makes him think "hard link"
12:39:25 Join kugel [0] (~kugel@
12:39:26 Quit kugel (Changing host)
12:39:26 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:45:13 Quit xavieran (Ping timeout: 265 seconds)
12:52:42 Join xavieran [0] (
12:55:45 Quit dfkt (Ping timeout: 260 seconds)
12:58:49 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:02:00 Join DerPapst [0] (
13:06:10*amiconn just got around to reading the monster thread on dev ml
13:11:51 Join ranmachan [0] (
13:14:01Stummigevaerts, ok, changed this
13:14:25 Quit kugel (Ping timeout: 260 seconds)
13:20:25 Join kugel [0] (~kugel@
13:20:26 Quit kugel (Changing host)
13:20:26 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:24:10 Join GeekShadow [0] (~Antoine@
13:24:10 Quit GeekShadow (Changing host)
13:24:10 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
13:27:31 Quit xavieran (Ping timeout: 240 seconds)
13:30:09 Quit casainho (Remote host closed the connection)
13:32:41 Join JdGord [0] (~jonno@
13:32:51 Join xavieran [0] (
13:36:48 Join sacredchao [0] (debian-tor@gateway/tor-sasl/nike)
13:36:52sacredchaoWhat program can I use to extract music files from an Apple-firmware iPod so that it reads the tags and actually names the files based on artist/album/number/title?
13:37:03sacredchaoI am on Debian and want to back up this music prior to Rockbox installation
13:39:14 Quit The_Pwny (Quit: If you can't laugh at yourself, make fun of other people.)
13:39:22 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
13:43:33kugelsacredchao: mp3tag will work (on Windows). but rockbox will find your music
13:43:48sacredchaoI am not on Windows
13:44:00sacredchaoI want to pull off the music on GNU/Linux and back it up
13:44:38kugeleasytag can possibly do it
13:44:56Bagderit can
13:46:38sacredchaoYeah but aren't the files on Apple-firmware iPods not only renamed to weird filenames, but put in incorrect folders?
13:46:58Bagderis funny-named folders too
13:48:10 Join kugel2 [0] (~kugel@
13:48:11 Quit kugel (Disconnected by services)
13:48:12 Quit kugel2 (Changing host)
13:48:12 Join kugel2 [0] (~kugel@rockbox/developer/kugel)
13:51:39 Quit GeekShadow (Quit: The cake is a lie !)
14:03:06sacredchaoSo there's no GNU/Linux application to fix this as files are copied off the iPod?
14:03:44Bagderdid you already try easytag?
14:04:13 Nick kugel2 is now known as kugel (~kugel@rockbox/developer/kugel)
14:11:35 Quit xavieran (Ping timeout: 272 seconds)
14:15:48 Quit DerPapst (Read error: Connection reset by peer)
14:16:42***Saving seen data "./dancer.seen"
14:17:48 Join DerPapst [0] (
14:19:59sacredchaoBagder: Found something better
14:20:03sacredchaothe answer: gtkpod
14:20:11 Quit kugel (Ping timeout: 240 seconds)
14:21:56 Join kugel [0] (~kugel@
14:21:57 Quit kugel (Changing host)
14:21:57 Join kugel [0] (~kugel@rockbox/developer/kugel)
14:31:26 Quit chattr (Quit: gone)
14:33:25 Quit kugel (Ping timeout: 240 seconds)
14:35:23 Quit JdGord (Quit: Bye)
14:36:25sacredchaoAlso Banshee but it wants to install mono on my system
14:36:29sacredchaofuck C#
14:37:15 Quit krazykit (Ping timeout: 240 seconds)
14:39:25 Join krazykit [0] (
14:42:06 Quit stoffel (Ping timeout: 265 seconds)
14:50:56 Join MethoS- [0] (~clemens@
15:02:40 Join marines [0] (
15:08:38 Quit krazykit (Quit: awe yeeeeeee)
15:22:19 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:22:24 Quit Bagder (Quit: Konversation terminated!)
15:32:05 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
15:33:07 Join chattr [0] (
15:36:11 Quit chattr (Client Quit)
15:37:51 Join Skaps [0] (~dpero@
15:37:59 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:38:01Skapshello! Skaps here.
15:40:02Skapsi have a sansa e260 and like, i want it to be able to play AVI files
15:40:26evilnick_BSkaps: Either convert them to mpeg or add in .avi playback ability
15:40:38Skapsi want to do the latter
15:40:44Skapsdo you have a link to get me started?
15:41:09evilnick_BAre you a programmer?
15:41:16Skapsby no means
15:42:03evilnick_BThat's not really an option then! It'd take a looong time to get up to speed on the code and then to work on adding support for .avi files.
15:42:18pixelmaisn't avi just a container anyway?
15:42:21evilnick_BRight now, your best option is to convert them
15:42:32Skapsdoes rockbox already have mpeg support?
15:42:35evilnick_BAnd pixelma is quite right, .avi isn't an actual codec
15:43:00Skapsso it should be very easy to convert then?
15:43:19AlexPSkaps: yes, mpeg1 and 2, see the manual
15:43:32Skapsthanks guys
15:46:37 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
15:47:56Skapsanyone know any legit avi to mpeg converters
15:48:55pixelmaour fine manual and wiki may
15:50:10 Join Strife89TX [0] (~cstrife89@
15:51:13Skapsyou are all so kind
16:01:14 Join Buschel [0] (
16:03:32Buschelsoap: any new results from your tests?
16:04:55 Quit utanapischti (Ping timeout: 240 seconds)
16:07:29 Join ruckus [0] (
16:07:46moospixelma: Hi, I adressed the cmments made yesterday, and here is the new patch:
16:08:41moosI reused a string that was here for recording only
16:10:53moosAnd I reused the current translations, that way, all language where the string was translated is reused
16:11:49 Quit eWill (Ping timeout: 240 seconds)
16:12:09moosif no one object, I will commit it later this day
16:13:23 Join eWill [0] (
16:13:30moosI'm planning to see if I can harmonise this with the properties plugin
16:13:46 Quit eWill (Client Quit)
16:14:29kugelmoos: can't one just change the id (I don't know)?
16:15:20moosI made this to not break compability and because it was a string depending on recording
16:15:34ruckushey everyone, i have a ipod video 30gig model (although with a 160gb drive) developed some weird problems recently, so I am wondering if anyone knows about them. 2 issues mainly, i would be playing music and it would randomly freeze. i can tell that the firmware is still working because i can scroll the song back and forth although there's no sound, and the song doesn't progress. also, often times i would play music and i wou
16:15:56ruckus*that developed
16:16:23evilnick_Bruckus: Try checking the disk
16:16:45***Saving seen data "./dancer.seen"
16:16:51evilnick_Bi.e. fsck or chkdsk
16:17:13mooskugel: I already updated all the current translation :)
16:18:05ruckusevilnick_B: fair enough, i haven't had the chance to do that.
16:18:40ruckusevilnick_B: could it be anything else ?
16:19:21evilnick_Bruckus: There are probably a few things that it could be, so best to rule things out one by one
16:21:53 Join b1uebrother [0] (~dom@
16:29:16 Join utanapischti [0] (~username@
16:29:24 Quit b1uebrother (Ping timeout: 265 seconds)
16:29:55 Quit dfkt (Read error: Connection reset by peer)
16:31:07 Join b1uebrother [0] (~dom@
16:32:20 Quit b1uebrother (Client Quit)
16:32:37 Join b1uebrother [0] (~dom@
16:32:44 Quit rasher (Quit: Lost terminal)
16:33:11 Join rasher [0] (
16:33:11 Quit rasher (Changing host)
16:33:11 Join rasher [0] (~rasher@rockbox/developer/rasher)
16:33:15 Join dfkt [0] (dfkt@unaffiliated/dfkt)
16:38:14Skapsevilnick_B, so ive converted an avi to an mpeg 2 and the playback on my sansa e260 isnt very smooth
16:38:21Skapswhat can i do to make it better
16:39:54evilnick_BDid you resize it to fit the display?
16:41:50 Join Buschel_ [0] (
16:42:16evilnick_BSkaps: Can you try out the sample video "Elephant's Dream" from
16:42:45 Quit Buschel (Ping timeout: 276 seconds)
16:42:47 Nick Buschel_ is now known as Buschel (
16:42:51 Join Skaps_ [0] (~dpero@
16:43:11Skaps_sorry i dcd
16:43:11Skaps_could you repeat anything you said
16:43:16evilnick_BSkaps_: Can you try out the sample video "Elephant's Dream" from
16:43:28 Join CaptainKewl [0] (
16:43:34Skaps_will do.
16:43:47evilnick_BThat will tell us whether it's your player or the video file that you converted that's making the playback choppy
16:44:52 Quit Skaps (Ping timeout: 245 seconds)
16:44:58 Nick Skaps_ is now known as Skaps (~dpero@
16:46:13Skapsill keep you updated.
16:46:16 Join bluebroth3r [0] (~dom@
16:49:06 Quit b1uebrother (Ping timeout: 260 seconds)
16:57:02 Join Buschel_ [0] (
16:57:31 Join fdinel [0] (
16:59:00 Quit Buschel (Ping timeout: 260 seconds)
16:59:15 Nick Buschel_ is now known as Buschel (
17:02:19Skapsevilnick_B, this video runs relatively smoothly.
17:03:50evilnick_BSkaps: So that points to the conversion process not being quite right. Check the settings that you used against the ones from the Wiki page and see if you can see any discrepencies
17:04:19Skapsrewatching the other video i converted, it isnt that bad actually
17:04:36Skapscompared to the one you gave me it runs at relatively the same smoothness
17:04:49Skapsthere are some laggy sections where the video seems to lose framerate
17:05:03ruckusevilnick_B: so i am running chkdsk right now, though i realized that my drive is fat32. we're talkinga bout 160gb, so is that good ?
17:05:05Skapsbut i guess that might be expected from a portable device of this caliber
17:05:56evilnick_Bruckus: Yeah, that will only affect how long the scan takes
17:06:02pixelmaSkaps: one thing to keep an eye on is overall bitrate
17:06:55Skapswhat kind of bitrate would run really well?
17:07:00Skapsanything under like.. 240?
17:07:09 Join chattr [0] (
17:07:32ruckusevilnick_B: what do you mean? performance wise it doesn't matter?
17:07:44 Quit TheLemonMan (Read error: Operation timed out)
17:08:03[Saint]copy the properties of the test files on the wiki
17:08:10[Saint]Skaps: ^
17:08:17 Quit Buschel (Ping timeout: 272 seconds)
17:08:20gevaertsruckus: are you asking if you should use something other than fat32?
17:08:23Skapsi have
17:08:41Skapsyou are referring to this page yes ?
17:08:45Skapsthe settings given on this one
17:09:08ruckusgevaerts: yes
17:09:28gevaertsruckus: rockbox only supports fat, so no
17:09:32evilnick_Bruckus: You HAVE to use fat32
17:10:19Skapsevilnick_B, Saint, i thank you
17:10:37 Quit sacredchao (Remote host closed the connection)
17:11:29Skapsmy bitrate on the VLC conversion tool is default at 800 kbps for Mpeg-2
17:11:33Skapsshould i decrease that
17:11:49Skapsi dont really know what a good standard bitrate is for video formats
17:11:50 Join sacredchao [0] (debian-tor@gateway/tor-sasl/nike)
17:12:27gevaertsI'd recommend experimenting a bit
17:13:21Skapsill try 400 and see if its watchable.
17:13:39evilnick_BSkaps: On that Wiki page it lists an average bitrate of 469 for the e200
17:13:47 Quit sacredchao (Remote host closed the connection)
17:13:48ruckusthank you gevaerts and evilnick_B. i have a feeling that i've already looked this up in the past, but i've forgotten it
17:13:53evilnick_B(at 4:3)
17:14:16Skapsand 354 for 16:9
17:14:51 Join WonTu [0] (
17:15:05 Part WonTu
17:15:11 Join sacredchao [0] (debian-tor@gateway/tor-sasl/nike)
17:18:39 Quit sacredchao (Remote host closed the connection)
17:19:51 Join sacredchao [0] (debian-tor@gateway/tor-sasl/nike)
17:20:02 Join Buschel [0] (
17:21:19 Quit Buschel (Client Quit)
17:21:40 Quit bluebroth3r (Quit: leaving)
17:22:32 Quit anewuser (Quit: Changing server...)
17:23:36 Join anewuser [0] (anewuser@unaffiliated/anewuser)
17:24:56 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
17:26:49 Quit sacredchao (Remote host closed the connection)
17:28:08 Join sacredchao [0] (debian-tor@gateway/tor-sasl/nike)
17:34:01 Quit Skaps (Read error: Connection reset by peer)
17:34:22 Join Skaps [0] (~dpero@
17:36:02Skapswell, i lowered the bitrate to 320 and it runs great
17:37:33 Quit sacredchao (Remote host closed the connection)
17:45:29 Quit DerPapst (Read error: Connection reset by peer)
17:46:00 Join DerPapst [0] (
17:51:17 Join stoffel [0] (
17:57:38 Quit petur (Quit: Page closed)
17:58:14 Join Keripo [0] (
18:00:50 Join Skaps_ [0] (~dpero@
18:04:37 Quit Skaps (Ping timeout: 265 seconds)
18:04:49 Nick Skaps_ is now known as Skaps (~dpero@
18:08:50evilnick_BSkaps: Excellent - now you know the settings, then you can convert whatever you like!
18:09:58Skapsthis 16 hour road trip should go down well.
18:16:48***Saving seen data "./dancer.seen"
18:18:25 Join casainho [0] (
18:24:32 Quit evilnick_B (Quit: Page closed)
18:25:14 Join toffe82 [0] (
18:27:07 Join b1uebrother [0] (~dom@
18:40:25kugelthere's still some dependency problem with lang.h
18:51:01 Join earcar [0] (
18:55:45 Join n1s [0] (~n1s@rockbox/developer/n1s)
18:57:27 Quit t0rc (Quit: Give someone code, help them with one project. Teach someone to code, help them rule the world.)
19:00:02 Part marines
19:02:08CIA-7New commit by alle (r28881): Fix a typo in comment
19:03:52CIA-7r28881 build result: All green
19:06:40 Join fml2 [0] (
19:07:30fml2Hello. Is .DS_Store on Mac (the itunes thing?) a file or a dir?
19:18:35 Quit toffe82 (Ping timeout: 265 seconds)
19:20:09 Join toffe82 [0] (
19:22:12 Quit casainho (Ping timeout: 240 seconds)
19:25:55 Join kadoban [0] (
19:28:56b1uebrotherfml2: .DS_Store is a file
19:33:20b1uebrotherhmm. tools/tools.make has a build rule for uclpack and tools/ucl/Makefile has too
19:37:08 Quit Guest17322 (Ping timeout: 272 seconds)
19:39:03 Quit Keripo (Quit: Leaving.)
19:40:21kugelgevaerts: shall I put the tagcache root patch on fs or commit now? Slasheri already looked at it a bit without complaint
19:42:50gevaertskugel: you probably answered this already at some point, but what's the impact on native targets?
19:43:15kugelno impact
19:43:28gevaertsthen I'd commit
19:44:13 Quit Strife89TX (Quit: Leaving)
19:51:00 Join madalu [0] (~user@unaffiliated/madalu)
19:53:12 Join bertrik [0] (
19:53:12 Quit bertrik (Changing host)
19:53:12 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:57:23 Join bluebroth3r [0] (~dom@
19:58:48 Join Horscht [0] (
19:58:48 Quit Horscht (Changing host)
19:58:48 Join Horscht [0] (~Horschti@xbmc/user/horscht)
19:59:05 Quit b1uebrother (Ping timeout: 255 seconds)
19:59:46CIA-7New commit by bluebrother (r28882): Set APPVERSION for libmkamsboot and libmkmpioboot. ...
20:00:00CIA-7New commit by bluebrother (r28883): Tweak Makefiles a bit to allow cross compiling Rockbox Utility. ...
20:02:10CIA-7r28882 build result: All green
20:02:23CIA-7New commit by kugel (r28884): RaaA: Improve tagcache search to make the database built. ...
20:02:35kugelgevaerts: still need to fix mod metadata parser though
20:03:04gevaertsAh, yes
20:03:24gevaertsI'd say that that's a different issue though
20:03:58CIA-7r28883 build result: All green
20:06:38CIA-7r28884 build result: All green
20:09:31 Join soap_ [0] (~soap@rockbox/staff/soap)
20:09:48 Quit soap (Disconnected by services)
20:09:49CIA-7New commit by alle (r28885): Add the new default dir to clean to the manual
20:09:51 Nick soap_ is now known as soap (~soap@rockbox/staff/soap)
20:10:20 Join soap_ [0] (
20:10:20 Quit soap_ (Changing host)
20:10:20 Join soap_ [0] (~soap@rockbox/staff/soap)
20:11:42soapBuschel, I've done some more tests (logging on right now to make a 76Mhz test build again) while driving today. With a stock build (force boosted to always run at 80 or not) the problems are happening repeatedly at the first flash access after the device gets warm.
20:11:49CIA-7r28885 build result: All green
20:13:35soapIn a 19 C vehicle the device will not crash ever. But much warmer it will crash the first flash access after the initial boot and buffer fill. Get it hot enough and the bootloader won't find rockbox.ipod.
20:13:52gevaertsAt flash access? Interesting...
20:13:57*gevaerts looks for grep
20:15:09soapSo let's call the three temps Cold (18C) Warm (?C) and Hot (?C). When warm Rockbox will fill the buffer (said second time) and start playback, but playback will be a mix of multiple tracks. When Hot the device freezes.
20:15:25 Quit fml2 (Quit: ChatZilla 0.9.86 [Firefox 3.6.13/20101206122310])
20:16:12soapWhen Warm said scrambled audio (fractional seconds of each track, jumping back and forth between tracks) lasts for a while (30 seconds?) before the eventual crash
20:16:14gevaertssoap: maraz reported similar (or at least also nano, heat, and storage related) issues back in Julu
20:16:53***Saving seen data "./dancer.seen"
20:16:54soapMy only test with the 76Mhz build was not under the same conditions as today's testing. I will repeat shortly.
20:17:21 Join JesusFreak316 [0] (
20:17:33soapBut it was under the same conditions which caused failure of a 80mhz build previously. I need to get back on the road, but should be able to report more in about an hour.
20:18:12soapgevaerts, look back to FS #7510's era. There were many reports in the FS and Forums and IRC at that time IIRC.
20:18:34soapI would not be suprised if, being as this appears to be a rare issue, some of the afflected people did not come back.
20:18:55gevaertsThat's certainly possible
20:23:37 Quit Skaps (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
20:24:56 Join [AndrewR] [0] (
20:25:41gevaertssoap: I think it would still be useful to know if it's DMA related for you as well
20:26:35soapHow would I test that?
20:26:40soapjust disable DMA?
20:27:04gevaertssoap: yes. According to that's what maraz did
20:27:10soap(ha! He says "just" as if it will be easy for him!)
20:27:27marazdisabling dma certainly fixed the problem for me
20:27:58soapI'll try that next.
20:28:19soapFWIW windows is unzipping my new build to said (ice cold) Nano at an amazing 12KB/s)
20:29:02soapIF it is aging flash going out of spec I'd be a bit shocked. This flash has very very very few writes on it.
20:29:04Stummi[0]=&sev[0]=&due[0]=&cat[0]=&status[0]=open&percent[0]=&reported[0]= can i commit this or should i await more comments for this?
20:29:18soapunless "aging" is truely time and not use dependent.
20:29:48 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
20:29:55marazsoap, my flash works just fine on OF.
20:30:07gevaertsmaraz: did you ever try a slightly lower boosted clock?
20:30:10marazrockbox with dma - problems. rockbox without dma - no problems.
20:30:18marazeven 24MHz.
20:30:50soapCurious. We'll see, the 76MHz test may have been a red herring.
20:32:02soaplarge files are trasfering fast. Lots of small files (full zip) are transfering very very slowly.
20:33:27 Join Feisar [0] (
20:33:53 Nick Feisar is now known as Guest28787 (
20:35:17kugelFS #11828 - Fix core mod metadata parser
20:35:26kugelwho's listening to mod files?
20:35:52kugelStummi: how does it handle themes?
20:35:56 Join GeekShadow [0] (~Antoine@
20:35:56 Quit GeekShadow (Changing host)
20:35:56 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
20:36:01 Join soap__ [0] (
20:36:15kugelthemes are only made at make install time
20:36:33Stummikugel: i don't checked this
20:36:52Stummi(and don't have any possibility to check this from here at the moment)
20:37:23 Quit soap (Ping timeout: 245 seconds)
20:37:53saratogathat reminds me, if someone feels like looking at the makefiles, having "make codecs" and "make plugins" not also include the themes, fonts, etc would be nice
20:38:11kugelsaratoga: they dont
20:38:40saratogathey did when i tried?
20:39:08 Quit stoffel (Ping timeout: 264 seconds)
20:39:09kugelonly make (full)install messes with themes and fonts
20:39:10Stummihm, i will wait until saturday or sunday. Then i am back to home
20:39:44kugelalso it's make rocks, not make plugins ;)
20:39:48Stummikugel: at the moment yes. It would be nice if somebody could source this out
20:40:12 Quit Guest28787 (Ping timeout: 260 seconds)
20:43:42saratogakugel: I just tried it and make codecs included wps, themes, a half dozen fonts, etc
20:44:04saratogathe majority of the zip size is actually things other then the codecs
20:44:24kugelmake codecs creates no zip
20:44:24gevaertssaratoga: just "make codecs"?
20:45:05saratogasorry, make codecs, then make 7zip
20:45:21kugelwell, make *zip also makes themes
20:45:46 Join afk [0] (~Dre@
20:45:55saratogais there someway to get a zip of just the codecs short of doing it myself?
20:47:36 Quit Dreamxtreme (Ping timeout: 265 seconds)
21:04:50 Join designate72 [0] (
21:10:35 Join Strife89 [0] (
21:13:26 Join krazykit [0] (~krazykit@
21:16:15 Join Keripo [0] (
21:23:31 Join soap [0] (
21:23:32 Quit soap (Changing host)
21:23:32 Join soap [0] (~soap@rockbox/staff/soap)
21:25:05[Saint]If it's not too much to ask, could somebody please roll me a build fot iPod Colour with this patch: applied?
21:25:31 Quit soap__ (Ping timeout: 240 seconds)
21:25:32[Saint]It would be very greatly appreciated, I am not able to build myself at the present point in time.
21:27:33[Saint]{And, if someone *does* roll me a build of SVN head, could the someone please include the .map file in the .zip file as well? Again, greatly appreciated - [St.])
21:34:22 Quit MethoS- (Remote host closed the connection)
21:35:09 Join wodz [0] (
21:36:34soapok, gevaerts, maraz, Buschel: I can not make the 76MHz build crash. I repeatedly got the ipod hot enough that touching the metal back to my cheek hurt.
21:36:40 Join Feisar [0] (
21:37:05soapIt didn't take 1/10th the effort to crash the 80MHz build, with or without frequency scaling.
21:37:06 Nick Feisar is now known as Guest33774 (
21:37:10marazsoap: from what i've gathered, it seems to be completely device specific at which point they'll exhibit symptoms.
21:37:37marazcould it be that apple never intended to use dma on the nano and bought chips that are more or less broken on their dma side?
21:37:52soapI'll test a no DMA build in a bit. I need to unload some equipment and run a few errands first.
21:37:56saratogaeasy way to tell: change the main build to 76MHz and see if people complain
21:38:44*The_Seven 's Christmas presents: and
21:38:45wodzI've got strange bug report regarding hd300. The guy claims that in rockbox anything above -15dB is clipped while in OF no clipping occurs with much higher gains. Do we had similar report for Gigabeat F or mrobe100 (using very similar codec)?
21:38:52soapOr, maraz, could Rockbox's DMA timings be off?
21:39:12marazwell, that's the other option.
21:39:21The_Seveni think this is ready for public alpha... people can now start playing around with their ipod classics :)
21:40:00[Saint]The_Seven: Congratulations, great work.
21:40:10saratogarockbox or iloader?
21:40:10n1sThe_Seven: very cool :)
21:40:24The_Seveniloader, allowing you to bootstrap rockbox :)
21:40:28[Saint]"The Masses" will be pleased, indeed, quite pleased.
21:40:37The_Sevenand it means that i can start actually porting rockbox now
21:40:47wodzthe only difference I spotted is that we set digital DAC gain to 0dB while OF set it to -0.5dB
21:41:26saratogaa lot of chips have separate gains for the DAC and headphone amps, maybe one is set too high?
21:42:08 Quit Guest33774 (Ping timeout: 264 seconds)
21:43:02wodzattenuating signal slightly in digital domain make sense IMO
21:46:41 Quit bluebroth3r (Quit: leaving)
21:46:58 Quit kadoban (Ping timeout: 245 seconds)
21:46:59[Saint]froggyman: <−− iLike Colour/Fuze/Photo
21:47:19[Saint]I shall do the "release" announcement this afternoon.
21:48:37[Saint]froggyman: Let me know if you find anything obviously wrong...or something you feel could be improved upon.
21:49:11froggymanwell, for starters I would prefer having battery as a percent and volume in dB
21:56:48[Saint]well...things that the OF does...that I'm not.
22:04:59*The_Seven wonders if it's possible to make rockbox compile without having an audio driver
22:05:44 Nick ej0rge_ is now known as ej0rge (
22:06:34kugelCaptainKewl: ping
22:07:31 Join mischasworld [0] (
22:10:53 Quit Keripo (Quit: Leaving.)
22:15:38 Join TheLemonMan [0] (
22:15:43 Quit mischasworld (Ping timeout: 245 seconds)
22:16:39 Quit ender` (Quit: The two aspiring triangle peaks arching into an abstracted ‘M’ typified the progressive leadership-minded outlook of the company. -- Motorola exec on their logo)
22:16:56***Saving seen data "./dancer.seen"
22:23:24 Join ender` [0] (
22:24:42 Quit factor (Read error: Connection reset by peer)
22:25:17 Quit afk (Quit: Don't follow me)
22:26:56 Join factor [0] (
22:27:15kugelamee2k: ping
22:27:37kugelany progress with your RaaA port?
22:28:10amee2knot yet, i've got all kinds of stuff getting in the way
22:28:20amee2kmostly xmas/holiday related
22:28:38kugelanother guy found that the SDL drawing is horrible slow (and I found in profiling that there are indeed a hundreds of thounds calls in it very shortly after starting)
22:28:43amee2ki'm also at my parent's place right now... good time to do some research
22:28:54amee2khence my question in -community
22:29:27amee2kthat would kinda explain why the interface is so sluggish
22:29:49kugelhe also made a patch to make it faster, so that it's usable on a n900; I bet it'll be useful for you too
22:29:55CaptainKewlkugel: pong
22:30:25kugelCaptainKewl: you listen to .mod files, right? can you have a look at FS #11828 please?
22:30:51 Join farlepet [0] (
22:31:12amee2ki'll try the patch right when i come back, which will be around something like january 3..5th
22:31:35 Quit soap (Read error: Connection reset by peer)
22:34:03kugelI should also get my mini2440 out of the drawer again to run the sdl app on it
22:35:26CaptainKewlkugel: I'll need some time to test it but at a glance it looks about right; I'll post my findings.
22:36:37 Join JdGordon [0] (3a6c66cc@gateway/web/freenode/ip.
22:38:48 Join Feisar [0] (
22:39:14 Nick Feisar is now known as Guest16638 (
22:44:00 Join Dreamxtreme [0] (~Dre@
22:45:33 Quit earcar (Quit: zzzzzz)
22:47:23 Quit Topy44 (Ping timeout: 245 seconds)
22:47:49 Join Topy44 [0] (
22:52:30 Join T44 [0] (
22:53:34 Join fml [0] (
22:54:14 Quit fml (Client Quit)
22:56:08 Quit Topy44 (Ping timeout: 245 seconds)
23:01:07 Quit farlepet (Quit: CGI:IRC (EOF))
23:01:52 Quit tchan (Quit: WeeChat 0.3.3-dev)
23:06:35 Join Clipper [0] (53182eb2@gateway/web/freenode/ip.
23:07:18ClipperDoes anyone know any issues of current stable release of rocbox with clip+ other than USB mode incompability?
23:08:06 Quit JdGordon (Ping timeout: 265 seconds)
23:09:05saratogakugel: nice catch with that mod parser
23:09:21saratogathats probably been the cause of more then a few db lock ups over the years
23:09:37saratogaClipper: check the wiki
23:10:29Clipperokay so
23:10:36 Join Buschel [0] (
23:10:54Clipperive downloaded rockbox utility 1.2.8
23:11:17Clipperconnected clip+ in MSC mode, launched roxbox untility and pressed autodetect button
23:11:23Clipperalways crash
23:13:17 Join JdGord [0] (~jonno@
23:13:19Clipperanyone else with clip+ experience this problem?
23:13:36gevaertsRockbox Utility crashes?
23:13:49Clipperyeah while autodetecting
23:14:16gevaertsI don't think that's common
23:14:40*gevaerts looks for domonoky or bluebrother
23:15:37Clippererror indicates problem with msvcrt.dll
23:16:10*gevaerts doesn't know
23:16:11Clipperwhich is supposed to be responsible for Visual C++
23:16:17Clippermaybe its windows problem?
23:16:23gevaertsCould be I guess
23:16:23Clipperim on XP SP3
23:16:36gevaertsThat should work
23:17:04pixelmareminds me of a similar problem with another Qt app (which the Rockbox Utility is also)
23:18:22pixelmathe solution there was to install a newer visual c++ runtime
23:18:34Clippersounds inspiring
23:18:36gevaertsvcredist something?
23:18:39pixelmado you know which you have installed?
23:18:52Clipperpixelma: nope, i dont know but im googling it at the moment
23:19:14gevaertsWhich visual c++ version is rockbox utility compiled with?
23:21:35wodzJdGord: FS #11829
23:21:40pixelmathe Utility ran on my XP SP3 laptop last time I checked, not sure I used autodetection though. For the other programme I needed this runtime:
23:21:40Tornegevaerts: none at all :)
23:21:42 Quit CaptainKewl (Ping timeout: 265 seconds)
23:21:51gevaertsah, right...
23:22:13gevaertspure mingw?
23:22:17Buschelsoap, maraz: your discussion about nano clock vs. DMA is interesting. IDE0_CFG (which has a special handling in the clock switching for the nano only) is changed by DMA code (ata_dma_set_mode)
23:22:17pixelmaso, I'm on the wrong track?
23:22:30pixelmawas just an idea
23:23:04kugelsaratoga: yea, it has a pretty serious buffer overflow problem
23:24:25Buschelsoap, maraz: can you check the upper 4 bits of IDE0_CFG in boosted/unboosted state with and withouth DMA?
23:24:29gevaertsBuschel: that's interesting...
23:25:07Tornethe PP ipods need certain minimal clock speeds for DMA to work
23:25:11Torneif that's relevant
23:25:42Clipperpixelma: installed, no change
23:26:36gevaertsTorne: ata_dma_set_mode() clears a bit in IDE0_CFG, which also cleared in ata_device_init(), *except* on the nano where it's explicitely set
23:26:47Tornenano1g, you mean?
23:26:59Torneok, i don't see why it would do that
23:27:03BuschelTorne: yes. but the issue is the other way round: too high clock seems to result in freezes on some nano's when using DMA access
23:27:04Tornebut i didn't write it. :)
23:27:11Torneask dreamlayers :)
23:27:21gevaertsIf the nano really is the same SoC as e.g. the video, there's almost no way that can be correct...
23:28:02gevaertsTorne: the weirdness is older
23:28:24gevaertsI suspect dreamlayers didn't see the nano special case elsewhere in the file
23:28:33pixelmaClipper: ok, sorry then I'm out of ideas. Maybe bluebrother or domonoky (the main Utility devs) could help but I don't know if they are really around). Can you go on with the installation when chosing the target and the drive letter manually?
23:28:36TorneAh, hm
23:29:48gevaertsThis would at least explain why DMA is involved
23:30:00gevaertsOr why it actually isn't :)
23:30:20 Join dovshap [0] (
23:30:54Buschelso, might just be some ifdef'ing needed...
23:31:10TorneRight.. hm
23:31:16Tornei suggest just do the same hack then :)
23:31:25Tornei don't think the IDE timing registers are well understood
23:31:27gevaertsOr try it at least :)
23:31:49Buschelwell, let's ask soap to do sp when he is back :)
23:31:59Buschelmy nano is 2G
23:32:09gevaertsand maraz I'd say
23:32:47gevaertsWe're still not 100% sure that they're having the same problem of course, but I think it's a good working hypothesis :)
23:33:26Buscheland it seems to be reproducible quite easily
23:33:40Clipperpixelma: well yes but i dont know if i can install
23:33:57Clipper7.0.2600.5701 (xpsp_sp3_qfe.081025-1544) <−− my version of msvcrt.dll
23:35:47pixelmait doesn't matter how the Utility found your device (manually or by autodetection) if that's what you think
23:37:52gevaertsTorne: the DMA code also sets some more unknown registers. How sure are we that just doing IDE0_CFG differently is safe?
23:38:15Torneno idea.
23:38:22Buschelsoap, maraz: as a first try I would just comment both IDE0_CFG lines in ata_dma_set_mode() in ata-pp5020-c
23:38:22Tornei have no idea how the PP-specific IDE stuff works
23:38:29Tornei just used dreamlayers' patch for all that
23:38:44 Quit JdGord (Read error: Connection reset by peer)
23:38:45Tornei only know about the non-SoC-specific parts of IDE i'm afraid
23:38:48gevaertsAnd how much faster is DMA on nano?
23:38:49 Join JdGord [0] (~jonno@
23:38:52Clipperpixelma: btw, <−−- reported by some sites as malware spreader
23:39:56Tornealso no idea
23:41:00TorneClipper: use MS's VC residistributable installer, don't download it from some reandom site
23:42:09Torne and the similar ones for the other versions of VC
23:43:29*Buschel just recognized he asked dreamlayers about the IDE0_CFG stuff in his DMA code
23:44:01*gevaerts saw that too
23:44:35Clipperpixelma: <−−−− is identified by microsoft as dongerous
23:44:49Buschelhe reengineered from iPod Video code. maybe the nano is a bit different in this area...
23:44:59gevaertsLooks like it
23:45:30 Quit ender` (Quit: A memorandum is written not to inform the reader but to protect the writer. -- Dean Acheson)
23:47:10ClipperTorne: it didnt help me but m getting latest NET distribution right now
23:47:15gevaertsClipper: then don't use that site. Nobody here recommended it...
23:47:25gevaerts.NET is irrelevant
23:47:38 Join JdGordon| [0] (
23:47:38 Quit JdGordon| (Changing host)
23:47:38 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
23:47:58 Quit JdGord (Quit: Bye)
23:48:51 Quit factor (Read error: Connection reset by peer)
23:48:52Clippergevaerts: i know that noone recommended that website, but as its one of top google picks related to msvcrt.dll phrase, im just trying to warn you
23:48:58Torneanyway, it's unlikely to actually be a problem with msvcrt, generally
23:49:04Clipperit looks like "i will help you" website while its a trap
23:49:06Torneit usually means the code's passed a bad pointer to something in libc
23:49:08Torneor similar
23:49:33 Join factor [0] (
23:50:17soap_gevaerts, you mention r15705. This heat-related "glitching" (let's just call it that as the exact nature of the glitching was not pinned down as well previously initially started before 15705.
23:50:29 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
23:51:15gevaertssoap_: yes. r15705 adds some nano-specific special cases, which are not taken into account in the ATA DMA patch
23:52:12soap_so go for starters with <Buschel> soap, maraz: as a first try I would just comment both IDE0_CFG lines in ata_dma_set_mode() in ata-pp5020-c ?
23:52:21 Nick soap_ is now known as soap (~soap@rockbox/staff/soap)
23:52:39*maraz eyes the task at hand
23:52:43*maraz decides to eat first
23:52:48Buschelsoap, maraz: can you check the upper 4 bits of IDE0_CFG in boosted/unboosted state with and withouth DMA?
23:53:15soapBuschel, I can, would it be easier to open a task to dump the finding in?
23:53:44BuschelI think so
23:55:14ClipperTorne: should i try to manually assign drive letters then?
23:55:43Torneif you want to get it installed, i suggest so :)
23:55:59Torneif you want to help us debug the issue with rbutil, then maybe not.. but i dunno what you can reasonably do to help there
23:56:04Torneunless you know how to use a windows debugger ;)
23:56:42Clipperi dont have it installed
23:56:55Clipperi was checking with some basic tools like dependency walker
23:57:04Clipperand msvcrt has no dependencies at all
23:57:15Torneif it was missing something it wouldn't start in the first place
23:57:26Torneand like i said, there's no particular reason to assume it's a problem with msvcrt
23:57:28 Quit bertrik (Ping timeout: 265 seconds)
23:57:40Tornethat's just hte C runtime, if any other code passes it a bad pointer it will crash in there
23:58:21ClipperAppName: rockboxutility.exe AppVer: ModName: msvcrt.dll ModVer: 7.0.2600.5701 Offset: 00037fd4
23:58:32Clipperfrom error log
23:58:44Torneyes; that doesn't tell anyone anything, unfortunately

Previous day | Next day