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 2008-10-28

00:02:06 Quit Hillshum (Remote closed the connection)
00:03:00 Join m0f0x [0] (i=m0f0x@gateway/tor/x-3a65d1c513a7cd5d)
00:06:03 Nick HBK- is now known as HBK (
00:06:11***Saving seen data "./dancer.seen"
00:08:19 Join hd [0] (
00:08:40 Quit HellDragon (Read error: 104 (Connection reset by peer))
00:08:46 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
00:09:17 Join Slack [0] (
00:09:26 Join aarcane [0] (
00:09:31 Quit jhMikeS (Nick collision from services.)
00:09:37 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
00:15:42 Quit ender` (" “That’s right, babe,” Cochrane chortled. “We’re not possessors, we’re just like dimensionally disadvantaged.” -- Peter F. H")
00:16:20lwizardlok how do i exit freedoom?
00:16:45bluebrotherby pressing the right key(s)
00:17:15lwizardlnm i had to switch on hold button to get the menu back
00:20:54 Join mc2739 [0] (
00:21:25 Join krazykit [0] (
00:27:10 Quit shotofadds ("Leaving")
00:30:33 Nick hd is now known as HellDragon (n=jd@Wikipedia/HellDragon)
00:30:59 Join Nibbl [0] (
00:31:25 Join allele2 [0] (
00:33:34 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
00:34:52allele2llorean, for a plugin creator (I don't think it would be a true compiler) would the plugin prettymuch iterate through the c file and convert each function into its binary counterpart and print it out into .rock file?
00:35:33 Quit lwizardl ()
00:35:37 Quit allele2 (Client Quit)
00:35:40 Join allele2 [0] (
00:36:19krazykitallele2, that's really just compiling it.
00:37:34 Nick allele2 is now known as allele (
00:39:50 Quit DataGhost (Nick collision from services.)
00:39:58 Join DataGhost [0] (n=dataghos@unaffiliated/dataghost)
00:40:08linuxstballele: Why would you want to anyway?
00:42:45 Join CaptainKewl [0] (
00:44:32 Quit HellDragon (Read error: 104 (Connection reset by peer))
00:45:07allelehehe, lots of reasons for me, overall I'd like to be able to compile on the go
00:45:26LloreanDo you often code on the player or something?
00:45:50krazykitallele, seems easier to just carry around a laptop.
00:46:24 Join HellDragon [0] (
00:46:43 Join dead3r [0] (
00:50:41alleleI know noone would code an entire plugin on their player, but if you make a plugin and notice a simple bug in it and know how to fix it why wait not just fix it on the player?
00:51:04 Join daurnimator [0] (
00:52:19 Quit m0f0x ("dang")
00:52:38Lloreanallele: It'd mean you'd need to keep an up-to-date copy of the Rockbox source on the player
00:52:45LloreanAnd your compile times would probably be longer than your battery life.
00:52:58allelefor a single plugin?
00:54:24LloreanDepends on the plugin, but some of our plugins are the longest-taking parts of the normal compile.
00:55:38allelealright well in any case the idea is now on the log and on topic and I have to go
00:55:40 Quit allele ("Java user signed off")
00:56:20 Quit bluebrother ("leaving")
00:59:24 Join daurnimas [0] (
01:00:45 Quit mc2739 ()
01:06:36 Part pixelma
01:07:07 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
01:07:46 Quit dead3r ()
01:13:09 Quit daurnimator (Connection timed out)
01:21:10 Quit daurnimas (Connection timed out)
01:21:44 Join daurnimator [0] (
01:23:46 Join fdinel [0] (
01:25:12 Join massiveH [0] (
01:26:02 Join dataangel [0] (n=chatzill@
01:27:00dataangelWhat's the best way for a plugin to get the time for N days from now? Not sure what combination of get_time and mktime to use, could just add seconds to result of mktime but I don't know if that's kosher
01:30:17 Join bughunter21 [0] (n=Jelle@
01:31:13 Quit PaulJam (".")
01:31:49 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
01:36:12 Quit culture (Read error: 110 (Connection timed out))
01:44:41 Join __indi_ [0] (
01:45:57__indi_and bye
01:46:00 Part __indi_
01:46:21 Quit bughunter2 (Read error: 110 (Connection timed out))
01:52:52 Quit miepchen^schlaf (Remote closed the connection)
01:59:42 Quit daurnimator (Connection timed out)
02:00:05 Join daurnimator [0] (
02:06:14***Saving seen data "./dancer.seen"
02:10:52 Quit saratoga ("CGI:IRC (EOF)")
02:16:29 Join nring [0] (
02:26:20 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
02:27:00 Quit faemir (Remote closed the connection)
02:27:05 Quit Thedjatclubrock ("Changing server")
02:30:05 Join m0f0x [0] (
02:31:32 Quit dataangel ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
02:48:24 Join daurnimas [0] (
02:49:54 Quit krazykit ("Connection reset by beer")
02:53:45 Quit daurnimator (Connection timed out)
02:58:18 Join Darksair [0] (n=user@
03:04:00 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
03:04:05 Join christophe [0] (n=chatzill@
03:05:02 Quit massiveH ("Leaving")
03:06:28 Quit daurnimas (Connection timed out)
03:06:45 Join daurnimator [0] (
03:07:14 Quit jhulst (Read error: 110 (Connection timed out))
03:07:26 Quit m0f0x ("sleep")
03:10:40 Join midkay_ [0] (
03:13:45 Nick fxb is now known as fxb__ (
03:15:33 Nick Darksair is now known as Darksair{away} (n=user@
03:15:39soapLlorean, are not these new "low mem" targets low mem _because_ they are flashed based and the manufactures have decided to forgo expensive RAM since it is of only marginal benefit on a flash-based player?
03:16:23LloreanBut in their case we need to fit the core *and* codec into RAM
03:16:31LloreanAssuming we drop the idea of plugins entirely.
03:16:42LloreanAs in, don't reserve the plugin buffer, I mean
03:16:46soapIf anything I think the low-mem flash targets argue for the settings-as-plugin direction, as they negate one of the strongest arguments against (disk hit).
03:16:53LloreanYes, they do.
03:17:15LloreanBut I can't imagine settings-as-plugin freeing up several MB of RAM.
03:17:40soapI assume the forking of settings-as-core and settings-as-plugin into two branches would be a nightmare to maintain?
03:17:43LloreanAs it stands we use 1MB for codec + codec RAM, iirc, plus the main binary and its in-memory structures.
03:18:08LloreanI think whether or not it's a nightmare to maintain comes second to whether it's necessary.
03:18:30LloreanAs it stands, we're probably going to *have* to have a "lite" or "lowmem" version of Rockbox for those targets.
03:18:43LloreanIt might be best to just pare down the features and have "lowmem only" targets and "lowmem or full" targets
03:19:13soapsettings-as-plugin-Mk-II, where HDD targets have a second plugin buffer for the settings plugin to be always loaded?
03:19:31soapand flash targets don't have said second plugin buffer?
03:22:55 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
03:24:00preglowsettings as plugin won't save much at all
03:24:02preglowjust the menu code
03:24:26JdGordonLlorean: nothing will free up several MB of RAM... in total we use 5mb on the worst targets... 2.5MB of tht is the codec, plugin buffers and the bin
03:24:32 Quit midkay (Read error: 113 (No route to host))
03:24:52LloreanJdGordon: Exactly. My point was, more or less, we need a more drastic solution for the very lowmem targets than paring down settings.
03:26:01LloreanObviously since they're flash we could cut most dircache functionality. We could probably remove database generation, requiring it to be PC-side for "rockbox lite". Drop the plugin buffer. Use a restricted subset of the codecs (only ones below a certain memory-use threshold). Tone down the UI a lot (since they're so far pretty limited screens, not too bad a problem here).
03:26:06LloreanMaybe even drop loadable fonts and bitmaps.
03:26:25 Join gregorovius [0] (
03:26:28soapHow much mem does database gen code take? That seems like perfect plugin fodder.
03:27:01Lloreansoap: Putting it in a plugin removes the ability to have it auto-update.
03:27:17preglowgod, i wish people would stop seeing stuffing things in a plugin as a solution
03:27:21preglowcore things should remain core
03:27:22soapHave said plugin lauch on boot.
03:27:37Lloreansoap: And USB. And file delete. And etc?
03:27:49Lloreanpreglow: Database *generation* is only dubiously core
03:27:54LloreanFor years it was external-app only.
03:28:13preglowwell, as it is now i can agree
03:28:20preglowhas to run on startup anyway
03:28:35soapI'm asking this not as someone who has the knowledge or skills to do it - but as someone wrestling with the philosophical question of it, preglow, so forgive me if I seem to be advocating one position over another.
03:28:35preglowbut i very much like the fact that rockbox can generate its own db
03:28:44soapHow else do you strip the core size down?
03:29:00gregoroviusHi. I'm currently using a nokia n800 as a music player, but i'm looking for something smaller and with more capacity. buying a used ipod and putting rockbox came to my mind, so I'd like to ask a couple things: does it show up as a usb mass device I can just copy music to, or do I have to put up with iStuff? and how's the battery life overall?
03:29:05preglowby optimizing and not including the kitchen sink
03:29:24 Part christophe
03:29:24Lloreangregorovius: no need for iTunes, and battery life is similar to what apple advertises.
03:29:38preglowbut yeah, i very much agree with llorean at least on the low-mem target side
03:29:42preglowwe need to do something drastic there
03:29:45Lloreanpreglow: Well, the hypothetical target has 384KB of RAM, so it's got to be more drastic than "optimizing"
03:29:50LloreanAh, right.
03:30:07preglowwell, putting anything in a plugin isn't a solution there
03:30:19preglowsince plugins will be _very_ restricted, if allowed at all, with so little ram
03:31:14LloreanHow much smaller are builds for the Player than for bitmap-targets?
03:31:18JdGordonLlorean: I really think that the 384KB target is irrelevant to the dissucssion.. MAJOR changes will have to happen for that port whatever happens, its the 8-16MB hard disk targets which are the only valid argument (imo) and they arnt being made anymore
03:31:30preglowour plugins really just aren't flexible enough for them to be used for anything else than just small programs and games
03:31:38LloreanJdGordon: You don't know that, at all.
03:31:40soap1,402,860 is current H300 RAM size. optimize that by an ungodly 50% and you're still large.
03:31:57JdGordonthe h300 has 32MB ram...
03:32:10soapmissing my point, JdGordon.
03:32:49soapMy point is a modest color-screen target consumes so much RAM that "optimizing" will never get 384 and thus pruning is the only option.
03:33:00preglowpruning is indeed the only option
03:33:21JdGordonand should be target specific when its closer to a mutliation...
03:33:38preglownot only that, but we'll have to do add alternate code for some parts, like reading files directly from flash, not to ram first
03:34:01*preglow shudders
03:34:03XavierGrpreglow: while I don't like the settings-in-a-plugin idea the main overhead of settings in size comes from the settings mechanism
03:34:04preglowi like my ram
03:34:28preglowXavierGr: are you sure? i'd be surprised if the menu code added more size than the settings handling itself for most settings
03:34:32ameyerwe talking about the clip here?
03:35:03*ameyer is too lazy to scroll up
03:35:24preglowameyer: not in particular, we're talking features and binary size, it seems
03:35:58XavierGrpreglow: well just a simple integer can add up to 300 bytes of code (setting function one liner)
03:36:11XavierGrpreglow: though I am not sure how much of this can be saved by a plugin
03:36:20preglowi don't really see any of this as a problem at all, i don't mind using more ram for features. the problem i see with this is that rockbox has a usability problem that'll only get worse with adding more features
03:36:26JdGordonthat part cant be moved to a plugin
03:36:34JdGordonits the menu's which can
03:36:51XavierGrah then I really don't know if it makes a difference
03:37:13JdGordonpreglow: adding more features doesnt make it worse... we've known for a long time we need a usability reowkr
03:37:23preglowi can see rockbox becoming one of those programs where i just see all of the settings then go "bah, i can't be bothered"
03:37:36ameyerwell, I assume Llorean's "hypothetical target" is the clip
03:37:43JdGordonyou think that doesnt happen already?
03:37:48preglowit does
03:37:48XavierGrpreglow: I find usability issues and features somewhat irrelevant. I don't think one contradicts the other. It is just that no usability improvemnts happened
03:37:49JdGordonameyer: yes
03:37:50preglowthat's my point
03:38:04JdGordonthen there is no loss by adding more
03:38:05preglowand i don't want to make it worse before it's gotten better
03:38:10JdGordonthe same people will ignore it anyway
03:38:12preglowwhat kind of thinking is that?
03:38:22preglow"it's bad, so we can just go on and make it worse"
03:38:30soaphow many "features" (tongue-in-cheek) are currently ifdef'd? Recording/Albumart/quick&pitchscreens/tagcache? Is that about it?
03:38:38XavierGrI still can't understand why a big list of selection can hinder usability
03:38:55XavierGreither way you will choose the setting you want to change, the others you just simply ignore
03:38:58JdGordonsoap: yeah, not many more than that
03:39:24soapdoes dircache consume squat if it is not enabled?
03:39:35JdGordonit will once my latest patch goes in
03:39:44preglowXavierGr: so you don't feel a difference in using a program that has tons of settings and one that has fewer, but manages to do what you want anyway?
03:39:45JdGordonbut atm its a couple k
03:39:49preglowi sure know what i like the best
03:40:37soapsounds like an argument for "non-core" settings to go non-core to a plugin. ;)
03:40:49JdGordonpreglow: the thing is, peple comeing to rockbox want more from their DAP, its not like PC apps where there is choice, its the option of the OF, or choice
03:40:54XavierGrpreglow: aren't you annoyed by a program that lacks the little thing you wanted to change?
03:41:13XavierGrpreglow: and yeah programs with tons of settings are on my top lists
03:41:16preglowJdGordon: oh, indeed, i don't want to pare rockbox down to an ipod
03:41:28XavierGrpreglow: most others are a major PITA
03:41:34preglowJdGordon: but i have a slight feeling that if we lower our standards on what goes in, rockbox will be a goddamned mess
03:41:41LloreanXavierGr: So you'd rather just put in every option you can imagine?
03:41:51preglowand in the end, also from a developers stand point
03:41:58LloreanRockbox already has some really messy features
03:42:10preglowtons of features that few people care about soon equates to tons of features developers don't know well
03:42:12XavierGrpreglow: some examples: Winamp vs Foobar, Windows Media Player vs Media Player classic
03:42:51XavierGrLlorean: Personally yes, but this is rather impractical for real use and a project such as rockbox (embedded limited resources etc)
03:42:55JdGordonok, its NOT black and white... bad features will stay out, badly done features will stay out untill they are done... its the (what you love to call) mediocre ones with maybe a small following which are being argued
03:43:21XavierGrLlorean: and yeah just like JdGordon said only quality-done features
03:43:45JdGordonpatches should only be rejected on technical reasons
03:43:50preglowit's kinda pointless arguing about this anyway
03:43:56preglowif there is a good feature, then it should go in
03:43:56XavierGrsomewhat true
03:43:58LloreanJdGordon: So we should accept *everything* that's not technically flawed?
03:44:03preglowwe'll argue about what is a good feature, tho
03:44:22preglowall i want is just for rockbox to become more usable and seem less like a bunch of tricks just thrown together
03:44:30preglowbut i can't argue that into not being
03:44:40preglowthat just needs someone skilled in user interfaces to come and do some work...
03:44:46preglowgod knows that person is not me
03:44:49JdGordonLlorean: not everything, but if there is a patch i want and you dont thats technically fine it shuold, and if there is one you want and I dont (and you are happy to commit it) then it should
03:45:12JdGordonpreglow: waiting for the messiah isnt helful
03:45:15LloreanXavierGr: not just well programmed. It needs to be a feature that, over all, improves the experience for people. That means any costs to future users (binsize is a cumulative, permanent cost, for example) should be outweighed by the overall benefits.
03:45:33LloreanJdGordon: So basically, as long as one person wants it, and it has no technical flaws, it goes in?
03:45:36LloreanThat's pretty ridiculous.
03:45:39XavierGrand Llorean yes I agree that Rockbox has some really messy features
03:45:48LloreanEvery feature has ONE person who wants it.
03:45:51LloreanOtherwise it'd never be coded.
03:46:04JdGordonit needs one commiter to wwant it... not 1 joe shmoe
03:46:14preglowJdGordon: making rockbox the biggest mess we can in the meanwhile also isn't helpful
03:46:16LloreanThe feature I want is "longer battery life"
03:46:19 Nick Darksair{away} is now known as Darksair (n=user@
03:46:20Lloreancan I remove features to get it?
03:46:32LloreanOr does it only work for adding code, rather than adding features?
03:46:46XavierGrLlorean: you know that this is affected insignificantly right now
03:46:54JdGordonthen we are back to squae one with nothing new ever going in
03:46:55LloreanXavierGr: Not on my Archos.
03:47:11XavierGrLlorean: true, that why they are tagged as the problem targets
03:47:20LloreanXavierGr: And whether it's significant or not is irrelevant to JdGordon's suggestion that "if one committer wants it, it should be done"
03:47:35JdGordonpreglow: small patches dont negativly affect things.. its the big ones which introduce mess
03:47:54JdGordonand its the samll ones which are being argued about
03:48:01LloreanI thought this was about no specific feature...
03:48:05preglowi'm not saying nothing should go in either, i'm just saying we shouldn't alter our standards too much
03:48:12LloreanThe ones MOST argued about are jpeg in the core, bitmap scaling, and multifont.
03:48:24*preglow wants jpeg in the core :>
03:48:25JdGordonnone of which has a working patch
03:48:29LloreanIn fact, other than XavierGr's recent patch, was there *any* other specific patch you had in mind with this complaint?
03:48:31JdGordonworking finished patch
03:49:12JdGordonsure, my last 3 commits, wps view mode, bassically any time i comimt something which isnt a maintenance commit
03:49:20 Quit Darksair (Remote closed the connection)
03:49:21LloreanJdGordon: And none of those were actually rejected?
03:49:29LloreanI mean, I'm failing to see the actual *problem*
03:49:34LloreanWhere is this list of wrongfully rejected patches?
03:49:40XavierGrfor me yes it is not about a specific feature, I just don't want to see the rejecting rate increasing
03:49:41JdGordonits the pointless arguments around it which pisses me off
03:49:52 Join Darksair [0] (n=user@
03:50:01LloreanJdGordon: Ah. "People complain to me, so I wanted to complain back in the mailing list."
03:50:23JdGordonXavierGr: commit your patch and see what happens
03:50:25LloreanSo, there's not actually any patches you feel were *rejected* wrongfully?
03:50:37JdGordoncustomizable quickscree
03:50:49XavierGrNo, Llorean it is about the ongoing habit making setting inclussion more stric
03:50:55LloreanXavierGr: How many patches do *you* think were rejected wrongfully?
03:51:04LloreanIf it's more strict, there *must* have been some rejections you felt should go in, right?
03:51:12preglowi don't really think we are getting more strict
03:51:23preglowperhaps as compared to the early years, but it's not the same project anymore
03:51:23XavierGrJdGordon: I can't commit my patch, I don't have access :P
03:51:28LloreanI think we just have more people, so more opinions get voiced.
03:51:31*JdGordon thought you did :p
03:51:32LloreanWe're LOUDER, not the same as "more strict"
03:52:02Llorean"more strict" would require that we actually create a barrier to inclusion of patches beyond simply saying "I don't like that patch"
03:53:20JdGordoni dont like that patch isnt low enough for you already?
03:53:39LloreanI'm confused by that sentence.
03:53:59JdGordonyou just said that the barrier to inclusion of patches beyond simply saying "I don't like that patch"
03:54:12LloreanI said that if we'd become more strict, there'd have to be an actual barrier.
03:54:27LloreanRight now, as far as I can tell, your complaint is "people keep telling me they don't like my patches."
03:54:32LloreanI don't see how that's in any form an actual "barrier"
03:55:06JdGordonok, fine, done
03:55:16XavierGrLlorean: Actually I keep reading comments of "Settings bloat" more frequently
03:55:45JdGordonLlorean: you have just implictly agreed that it only needs one commiter to like a patch for it to go in
03:55:46XavierGrthat is forming a negative filtering for further features
03:57:26preglowhow did he agree to that?
03:57:27LloreanJdGordon: I never said it requires more than one.
03:57:41LloreanAll I said was, features shouldn't go in *just because* one person wants them.
03:58:01JdGordonwhich is not the same.. how?
03:58:23LloreanDoom *requires* 2MB of RAM to run. Just because you have 2MB of RAM, should you run doom?
03:58:33XavierGrpreglow: indeed it's not the same project anymore. In the old days I remember it more user friendly (and I don't mean in software)
03:58:53LloreanOne person can commit any features. The question of whether they should is not the same question as whether they're capable of doing it.
04:02:14preglowXavierGr: which comes from not involving so many people
04:02:27preglowthings get more complicated with more people onboard
04:02:34preglowthat happens to absolutely every project
04:02:47LloreanAs the number of people involved increases, the probability of voiced dissenting opinions approaches 1.
04:03:09XavierGrpreglow: true, I remember it had much more of a family feeling thing :)
04:03:25preglowyep, and i actually prefer smaller projects for just that reason
04:03:52preglowbut there's nothing to be done about that, rockbox has gotten big, and that is good
04:04:09JdGordonthen pre commit talks should start something like "I'm going to commit FS#... I feel it is in line with the projects goals, if you have technical objections voice them..."
04:04:21XavierGrpreglow: of course and I hope further for its best
04:05:05LloreanJdGordon: Just saying "I feel it is in line with the project goals" doesn't immediately make you right about it, and people will still voice disagreement.
04:05:16LloreanTechnical objections are *not* the only objections valid on a patch.
04:05:28XavierGrthough I am deeply conserned with its future. Devices tend to increasingly merge to phones these days :(
04:05:46preglowXavierGr: well, then we'll code on phones
04:05:58LloreanPatches should be discussed on the mailing list more often.
04:06:03preglowthe only thing that could threaten rockbox' future is if everything ran on high-powered generic hardware
04:06:06XavierGrrockbox might as well have a (distant) future only as an application on those devices
04:06:08preglowand that won't happen any time soon
04:06:08LloreanYou get more voices, and a better way to see how many devs feel what way about it.
04:06:17***Saving seen data "./dancer.seen"
04:06:30preglowXavierGr: well, rockbox as an app isn't so horrible, is it?
04:06:51XavierGrLlorean: I agree that patches should be discussed more on the mailing list, though sometimes there is complete absence of opinions there
04:06:54JdGordonLlorean: there is still the magic number of how many objections does it take to concede?
04:07:02JdGordon1? 2? 6?
04:07:14XavierGrpreglow: of course not, it just has a different feeling to it.
04:07:23preglowheh, sure
04:07:30LloreanJdGordon: No. Just like there's no magic number for how many people must like it before it goes in...
04:07:42preglowXavierGr: i sure know i'd love rockbox as an app on my mobile, perhaps then i'd use it for music
04:07:42LloreanI've never denied that in the end it's one person's judgment call.
04:08:12JdGordonpreglow: na, I still wouldnt use my phone for a DAP... its battery life is more imporant to me than music
04:08:14LloreanBut all I can see is that you're complaining about being complained at, since nobody's actually *rejected* anything you can complain about.
04:08:17XavierGrpreglow: truly that would be amazing to have, just like those linux-motorola mobile phone users
04:08:56XavierGrLlorean: you misunderstand, it is just worries of increasing settings censorship
04:08:57preglowJdGordon: well sure, depending on that of course
04:09:05preglowif i felt i needed music more than contacting people, i'd play music
04:09:10LloreanXavierGr: "Censorship"?!
04:09:21JdGordonXavierGr: no he gets it... we have moved off a bit from the origional argument
04:09:23XavierGrI mean acceptance
04:09:34preglowXavierGr: well, there's no need to worry about that; there's no sign that it's happening
04:09:35LloreanWhat, *exactly* is the poitn?
04:09:39LloreanNobody has rejected any features.
04:09:44LloreanOnly like, two features have been reverted ever.
04:09:47JdGordoncustomizable quickscreen
04:10:08LloreanSo the point is "People keep complaining, and because of that, I feel intimidated and choose not to commit features even though I'm sure they're good, but not sure enough to overrule the complainers"?
04:10:24JdGordonyou tihnk thats not a good attiude to have?
04:11:01JdGordonah fuck this, im hungry, i will loosen my attitude to other's opinions
04:11:05LloreanI think that if you're sure the feature is the right way to go, you'd have enough confidence to say "Look, everyone's just voicing opinions, I know nobody will always agree, but I think this feature is the right thing for Rockbox, and it's implemented in the most positive way right now"
04:11:19XavierGrLlorean: my worry is that a negative consensus on settings might be ensured
04:11:36LloreanHonestly, if I were you, I'd commit the version of XavierGr's patch that uses a setting, and be done with it, rather than creating topics in the mailing list.
04:11:40preglowi desperately need to sleep
04:11:42preglownightie all
04:11:58LloreanXavierGr: So people shouldn't voice opinions? *That* would be censorship.
04:12:05LloreanBecause that's all the topic is here: People voice a lot of opinions.
04:12:22JdGordonnow my problem is if its commited everyone will yell that i did it out of spitw
04:12:25XavierGrLlorean: if he commited just like that there would be a lot of flame
04:12:43LloreanXavierGr: And?
04:13:05LloreanIf you're sure the feature is *right*, why do you actually care about the flame?
04:13:21JdGordonnoone is ever sure of anything
04:13:33LloreanI would disagree.
04:13:41LloreanMaybe not "sure" but "confident" at least.
04:13:49XavierGrLlorean: I thought commiting patches was a group process, and if there are serious objections it can't be commited
04:14:04LloreanXavierGr: If that were true, we'd only have one or two committers, and a vote.
04:14:31 Join echelon [0] (
04:14:42LloreanWhen you're given commit access, that's a card saying "The project trusts your judgment, so try to use it wisely." That includes judging when you decide people are just being vocal, but they're vocalizing against the real best interests of the software.
04:15:07echelonfor some reason my sansa isn't booting into its original firmware when i hook it up to usb
04:15:12echelonso i can't mount it
04:15:25LloreanIf it becomes a real problem, either way, then we stop and discuss where we got off track. But if you really let loud voices stop you, nothing ever really happen, but you (the person listening to them) is the one really stopping progress, not the loud folk.
04:15:38XavierGrLlorean: well I've seen len0xs commits getting reversed, that wasn't good for the project either. The man got angry and left. Of course maybe it wasn't right to commit in the first place, it gets complicated
04:16:10echelondid i come at a bad time?
04:16:11*JdGordon doesnt even recoginse that nick :p
04:16:16JdGordonechelon: which sansa?
04:16:41echelonc250 ver 1
04:16:47JdGordonechelon: well, either way, hold left while its booting, maybe usb detection isnt working brilliantly
04:17:00LloreanXavierGr: Changes in *how* things work are probably something that should be discussed, since you're taking away old behaviour
04:17:12*ameyer guesses echelon has one of "those" builds that freezes on usb plugin
04:17:13Lloreanyou don't *have* to, but part of having your judgment trusted is a willingness to try to discuss things first.
04:17:21echelonit hasn't happened before
04:17:42echeloncool, thanks :)
04:18:26 Join blkhawk- [0] (
04:19:05echelonwhat gui library does the sansa use for its apps
04:19:24LloreanXavierGr: And frankly, len0x's attitude was really bad.
04:19:35Llorean"I'm a new young developer, and the old guys don't want to lose control" kinda stuff.
04:19:42XavierGrLlorean: I agree on that
04:19:48Lloreanechelon: What do you mean?
04:19:54JdGordonhow long ago was this?
04:19:55XavierGrI am just saying it was a bad experience
04:20:07LloreanJdGordon: 2005, google "len0x"
04:20:36LloreanXavierGr: That was a case of him putting his personal use preferences in, in a way that removed the existing use preferences though.
04:20:56LloreanYour patch, for example, doesn't remove anything existing (at least as a setting)
04:21:01JdGordonhehe first hit is about settings
04:21:05LloreanI'd probably revert it if you hard coded "+++" as the string, for example.
04:21:20LloreanIf it's going to be hard-coded, we need a consensus on what the new value is.
04:21:35JdGordonbut too many settings are bad (apparently)
04:21:39LloreanIf it's going to be a setting, on the other hand, you just need to be confident that it really improves usability more than it hurts usability.
04:21:44XavierGrI agree
04:22:07LloreanJdGordon: Too many are. That *doesn't* mean *all* new settings are.
04:22:10echelonLlorean: SDL?
04:22:15LloreanPersonally, I think this one is on the lower end of "acceptable"
04:22:37Lloreanechelon: The simulators use SDL. If you mean inside Rockbox, there's nothing but custom code.
04:22:56echelonnot even ncurses?
04:23:01JdGordonechelon: we have out or graphics library which you need to use, mostly just letting you draw lines and rectangles
04:23:03Lloreanechelon: Rockbox isn't Linux.
04:23:11echelonoh.. what is it?
04:23:13 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
04:23:20echeloni see
04:23:29LloreanXavierGr, JdGordon: By "lower end" I mean "it's acceptable for commit, but it's a fringe feature and is expected to incite debate"
04:23:35Lloreanechelon: Rockbox is Rockbox.
04:23:46LloreanIt's a custom, from scratch system.
04:23:55echelonnice :)
04:24:17JdGordonechelon: have a look at some of the plugins... we have everything you need.. you bring up ncurses, we have a list and menu api which is just as easy
04:24:49ameyerrockbox, in some cases, does "borrow" code from certain linux distributions, though
04:25:04Lloreanameyer: Rockbox also borrows code from a variety of codecs, etc.
04:25:25LloreanBut Rockbox is a very, very far stretch from *being* Linux or a UNIX-like OS in general
04:25:27ameyerand doom...
04:25:59 Quit Darksair ("Emacs = ESC-Meta-Alt-Ctrl-Shift")
04:26:05saratogawhat is the low mem target people were arguing about before?
04:26:07echelonyou're right.. it's not reading the usb
04:26:12echelonwhat's wrong with it?
04:26:14Lloreansaratoga: Clip. And not arguing about that one.
04:26:20XavierGrLlorean: At start I didn't have any hopes about it, it started as a self-hack (that's why hardcoded). Then other people got interested and more customization was preffered. In the end it ended up as a setting.
04:26:42saratogathe Clips got 2.3MB it seems, which ought to be enough for a Black+white target
04:26:45LloreanXavierGr: Well, you really shouldn't post to the tracker unless you're submitting it toward commit. That's what the tracker's for.
04:26:51saratogawell BW flash target
04:26:55JdGordonsaratoga: arg, that makes it worse actually...
04:26:56Lloreansaratoga: Maybe I'm confused. Which one was 384KB?
04:27:28saratogawe thought the clip had 320KB until last week when funman found another 2MB hidden on it, so maybe thats what you're thinking of?
04:27:37Lloreanmaybe that's it.
04:27:46ameyer384KB ought to be enough RAM for anybody...
04:28:08LloreanXavierGr: Anyway, I've been on the receiving end of wanting to commit a patch that a *lot* of people (in fact possibly a significant percentage of our current user-base for the affected target at the time) were in vocal objection to.
04:28:16saratogaI don't think we're ever likely to get a target with less then 2MB ram now that the ifp port is dead
04:28:21JdGordonLlorean: the tracker isnt really only for patches which should go in... there is no reason why a patch cant go there and sit waiting for support, (unless the author disappears when it comes to commit time)
04:28:29LloreanSo I know what it's like to hear people complaining. But you (or your designated committer, JdGordon) just need to make up your mind whether the patch is "right" for rockbox or not, and then do it.
04:28:43JdGordonLlorean: which patch was that?
04:28:45echelonJdGordon: so what can i do if usb detection isn't working?
04:28:49LloreanJdGordon: I said "toward" commit.
04:28:51echelonnothing shows up in dmesg
04:29:02saratogaare you in the OF?
04:29:05LloreanJdGordon: As in, any patch that's working toward commit, even if it's not ready *yet*, but not patches that are just intended for the self.
04:29:11LloreanJdGordon: Sansa e200 keymap
04:29:13echelonthe sansa knows it's connected since it's showing the charging indicator
04:29:24JdGordonLlorean: oh yes, I remember
04:29:37JdGordonechelon: your on ubuntu?
04:29:48saratogaechelon: that was directed at you
04:29:57LloreanJdGordon: Again it was a very subjective argument. But I was pretty darn sure the right way, in the long term, would be to have the "Rockbox" keymap rather than the "Sansa" keymap, for overall usability.
04:30:23echelonsaratoga: OF?
04:30:29saratogaoriginal firmware
04:30:54echeloni am now
04:31:28LloreanJdGordon: Anyway, I'm off for about two hours at least. I'm sure this isn't the last I'll hear of this anyawy.
04:31:33JdGordonechelon: try a different port on your computer, check dmesg for usb resets... if your in the OF and no usb then your on your own :(
04:31:46echelonk :\
04:32:03saratogareading the logs, I think complaining about settings wasting memory is rather silly when the codecs probably waste a good 0.5MB, maybe more
04:32:04JdGordonLlorean: na, im done
04:32:30JdGordonsaratoga: 1MB... but they are the whole point of rockbox so its sort of valid
04:32:42echelonshould USB mode be auto-detect or MSC?
04:32:51JdGordonbut yeah, the big part of adding settings is stuck in the core
04:32:51saratogaI don't think all of the 1MB is wasted, since we need some ram for codecs
04:33:45 Quit blkhawk (Read error: 110 (Connection timed out))
04:34:09echeloni'll try a reboot
04:34:25 Nick blkhawk- is now known as blkhawk (
04:36:01saratogalooking online there are embedded AAC-HE/PS decoders that use less then 200KB of RAM, verses the 1MB ours uses
04:36:31ameyerprobably not open-source, though
04:36:57saratogaof course not, otherwise we wouldn't need faad
04:37:41ameyerheck, open-source doesn't even matter that much
04:37:57ameyermore like GPLv2 compatible
04:41:08echelonJdGordon: the reboot worked
04:41:20JdGordonoh good :)
04:41:29echelonwonder what the issue was
04:41:54 Join midkay [0] (n=midkay@rockbox/developer/midkay)
04:45:09gregoroviusif I were to buy a classic-line ipod solely to use with rockbox, would it make sense to get a video over a 5g?
04:45:42gregoroviusis there any difference?
04:45:47cool_walking_Video = 5G
04:46:23cool_walking_The 6G (Classic) is the one you don't want, because Rockbox doesn't run on it.
04:46:48gregoroviuscool_walking_: sorry, I meant to say 5g vs 4g
04:47:03gregoroviusipod video vs ipod photo or color
04:47:14 Quit cool_walking_ (" ajax IRC Client")
04:47:25BHSPitMonkey5G has a nice qvga screen
04:47:49BHSPitMonkey4G Photo is less pixelles
04:47:54 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
04:48:39gregoroviusyeah, but besides from that? can the 4g play video smoothly w/ rockbox?
04:51:25cool_walking_ says 4g gets faster video playback.
04:51:35 Quit midkay_ (Read error: 110 (Connection timed out))
04:52:10Unhelpfulthe video certainly can't, in rockbox... but the OF, using the hardware decoder, is probably your best bet for video
04:53:13Unhelpfulor a gigabeat. the ipod line are not the very best for rockbox video.
05:00:19gregoroviushm, it seems video on the 5g isn't quite top notch
05:02:24 Quit daurnimator (Connection timed out)
05:02:44 Join daurnimator [0] (
05:02:55Unhelpfulit's only good in the OF, there's a hardware decoder that's quite nice, as i understand, but not documented, so only apple gets to use it.
05:04:11gregoroviusI see. in any case, video wouldn't be my main use, and I guess I could reboot to the OF in those cases
05:05:04 Join daurnimas [0] (
05:05:43Unhelpfulfor overall processing muscle, the gigabeat is probably your best bet. it'll do much better if you want to use the more cpu intensive audio codecs.
05:06:39soapcool_walking_, that is a bit misinformative.
05:06:54 Quit aarcane ("Leaving")
05:07:52 Join ^thehatsrule^ [0] (n=none@unaffiliated/thehatsrule/x-873637)
05:07:55gregoroviusUnhelpful: I don't think I can get a used gigabeat at a decent price... or at all, I only see a 10gb on craigslist here
05:08:00soapThe 4G and the 5G have almost exactly the same playback speed _per pixel_, the 4G just has less pixels. load the 4G resolution video on the 5G and you'll get the same speed.
05:08:03 Join Darksair [0] (n=user@
05:08:09cool_walking_Okay but I didn't want to bother saying "is the same speed but has a smaller screen so blah blah"
05:08:55soapso you'd rather say something misleading than something clear?
05:09:04cool_walking_Are you scorche?
05:09:45cool_walking_I said video playback, not video decoding/drawing speed.
05:09:48soapI can get you a 10GB Gigabeat F for $50 or a 40GB one for $100. Neither of those are great prices. toffe might have better prices.
05:10:07gregoroviusi'm not in the us of a
05:10:17soapand that's still wrong. Comparing like to like and they are alike.
05:10:37soapgregorovius, ok, add $10 to those prices, assuming central europe.
05:11:11cool_walking_so I should have added a "at full screen for each target" qualifier?
05:11:40gregoroviussouth america, actually
05:11:53soapI've never shipped there, hold on a min.
05:13:12soapvaries greatly. $10 for brazil, quite a bit more for Argentina.
05:14:33 Quit saratoga ("CGI:IRC (EOF)")
05:15:39soapBut back to iPods as Rockbox targets. The iPod G4 has probably the worst battery life (maybe except for the MiniG1???)
05:16:18soapThe G5 60/80GB variants the best. (more RAM and bigger batteries than the 30GB variants)
05:16:37gregoroviusargentina, and any case, soap, I don't trust the mail to get anything over $0.10 to me
05:17:12gregoroviusI was thinking of getting a 5g 60 or 80gb, yes
05:20:34 Quit daurnimator (Connection timed out)
05:20:35 Quit Zarggg ()
05:21:41 Quit FOAD (Read error: 110 (Connection timed out))
05:22:34 Part echelon
05:29:08 Quit kachna (Read error: 104 (Connection reset by peer))
05:29:23 Join kachna [0] (
05:34:32 Quit markun (Read error: 104 (Connection reset by peer))
05:34:42 Join markun [50] (n=markun@rockbox/developer/markun)
05:42:15 Quit gregorovius ()
05:44:26^thehatsrule^hi, would anyone know if the sansa e260 with the product id/model SDMX4-4096-C70 is compatible with rockbox?
05:44:46 Quit Horscht ("I am root. If you see me laughing, you better have a backup")
05:46:45advcomp2019^thehatsrule^, nope.. you need to look at the firmware version for sure
05:51:25 Quit nring (Remote closed the connection)
05:54:27^thehatsrule^alright, thanks
05:56:50 Quit XavierGr ()
06:06:19***Saving seen data "./dancer.seen"
06:06:37 Quit ghen ("leaving")
06:07:48 Quit Darksair (Read error: 104 (Connection reset by peer))
06:08:27 Join Darksair [0] (n=user@
06:31:28 Join AndyIL [0] (i=AndyI@
06:36:54JdGordonis a 3.3KB ram saving worth making buffer_alloc() harder to make slightly more dynamic?
06:37:49JdGordonless actually... 832 bytes even
06:44:25 Join reacocard_ [0] (n=reacocar@WL-112.CINE.HMC.Edu)
06:44:55 Quit AndyI (Read error: 110 (Connection timed out))
06:56:09 Join puzzles [0] (n=dan@xmms2/developer/puzzles)
06:56:46 Join HBK- [0] (
06:59:30 Join m0f0x [0] (
06:59:38puzzlesi'm having trouble with the (obviously experimental) video support on an ipod video. it seems no matter what combination of mencoder flags i use, i can't get audio.
06:59:56 Nick Darksair is now known as Darksair{away} (n=user@
07:00:15puzzlesis there some magic set of mencoder flags? or does mencoder do some weird encoding that the mpegplugin can't understand?
07:03:22cool_walking_puzzles, does the converted file work on your PC?
07:04:34cool_walking_and.. there is an mencoder example on the PluginMpegplayer wiki page.
07:04:48 Quit CaptainKewl (Read error: 110 (Connection timed out))
07:05:20puzzlescool_walking_: yes it does. and fwiw, i've been using that example and varying it
07:06:45puzzlesthe particular files i was reencoding actually worked, but played very choppy and with the wrong aspect ratio. so i even tried -oac copy and just reencoded the video, but still no audio
07:13:03 Nick Darksair{away} is now known as Darksair (n=user@
07:13:07LloreanWait, by "worked" you mean when you didn't use -oac copy it had audio?
07:14:00 Quit HBK (Read error: 110 (Connection timed out))
07:15:29puzzlesLlorean: yes
07:15:50puzzlesLlorean: oh sorry, no. i misread.
07:16:05puzzlesthe original file "worked" in that the audio played fine
07:16:31puzzlesbut no matter how i tried to reencode it, the audio always failed, even if mplayer played it fine
07:16:32LloreanBut the output file, using the original line (not -oac copy) plays fine on your computer, but not on the device?
07:17:10LloreanAnd you're definitely testing the output file, not the original?
07:17:47puzzlesi've gone through dozens of combinations by now and kept comparing to the original
07:17:55LloreanWell, if you can produce a small sample that's legally distributable of a file that should work, you should post it to a bug report.
07:18:11LloreanThe line in the wiki should work, and the file you get should play the same on your PC and on the player.
07:20:43puzzlesi'll see what i can cook up
07:24:05 Quit m0f0x ("dang")
07:35:09 Quit Slack (Read error: 60 (Operation timed out))
07:38:21 Nick fxb__ is now known as fxb (
07:38:32puzzlesnow to just wait an hour while a small chunk of elephant's dream is encoded ;)
07:40:25amiconnsoap: If you load a G4 resolution (color/photo) video on a G5, it will play *faster* than on a Color/Photo with SVN code
07:40:49amiconnThe G5 has asm optimised YUV->RGB conversion, and faster LCD writing
07:41:11*amiconn has some code in the works to change this and bring the Color/Photo up to speed
07:41:11cool_walking_Thanks amiconn, now I'm even wronger...
07:42:42 Join spiorf [0] (
07:43:45 Quit Nibbl (Read error: 60 (Operation timed out))
07:46:08 Nick fxb is now known as fxb__ (
07:46:46puzzlesLlorean: the elephant's dream from the wiki isn't even playing audio on the ipod!
07:48:55advcomp2019i though elephant's dream encoded to work already
07:49:29puzzlesit looks ok, aside from being choppy
07:49:34puzzlesbut no audio
07:49:50puzzles...and now some videos that worked before won't give me any audio either... wtf
07:50:50 Quit BHSPitMonkey (Remote closed the connection)
07:51:01puzzlesi think i figured this out
07:51:50advcomp2019did you put elephant's dream on to your player without reencoding? because the elephant's dream on the wiki should work without reencoding it if you download the one for the ipod
07:51:55puzzlesit seems there's no sound if you don't first play some audio in the normal rockbox interface
07:52:04puzzlesthen start a video in the mpeg plugin
07:52:44puzzlesall this time i was testing, i paused the song on startup because it was annoying, then i tried to play a video
07:53:26puzzlesand sure enough... my futurama episode i've been trying to encode this whole time works fine :D
07:54:34amiconnpuzzles: You don't need to play some audio before video, I think the problem is that you *paused* audio
07:54:41puzzlesi did
07:54:53amiconnIf you would have *stopped* it, mpegplayer would have worked
07:55:05puzzlesstop is unknown to ipods :P
07:55:06LloreanThat's certainly a bug, though.
07:55:09Lloreanpuzzles: No, it's not.
07:55:22amiconnLooks like a bug in mpegplayer - it needs to unpause pcm if necessary
07:55:40Lloreanpuzzles: Yup. It's covered in the manual. Long-press of pause, but not long enough to shut down.
07:55:43puzzlesmm. i'll go file a bug report if someone wants to point me to the tracker
07:55:54LloreanThere's a "bugs" link everywhere on the site
07:55:56puzzlesLlorean: ah i didn't realize that was different.
07:56:04Lloreanpuzzles: It's covered in the manual.
07:56:37amiconnAnother bug due to our dreaded pcm pause problem (it also breaks the voice UI)
07:56:39 Join bertrik [0] (
07:56:45 Join ajonat [0] (n=ajonat@
07:57:17Lloreanamiconn: At least it should be a fairly easy fix for mpegplayer though, right?
07:57:37amiconnplugin_get_audio_buffer() stops playback if necessay - that one should also unpause if paused
07:57:46JdGordonamiconn: do you know how pcm is supposed to work?
07:57:46JdGordonpcmbuf that is
07:58:10 Quit AndyIL (Read error: 104 (Connection reset by peer))
07:58:53 Nick reacocard_ is now known as reacocard (n=reacocar@WL-112.CINE.HMC.Edu)
08:01:18 Join AndyI [0] (i=AndyI@
08:02:56 Quit AndyI (Client Quit)
08:06:23***Saving seen data "./dancer.seen"
08:06:27 Nick bughunter21 is now known as bughunter2 (n=Jelle@
08:09:00JdGordonif dircache is compiled but disabled it currently wastes ~800 bytes stack, 1 thread and however big the event queue is... stopping the thread being created is easy, sis it worth buffer_Alloc()ing the two buffers? (or at least the stack space) to reclain some RAM?
08:09:35advcomp2019amiconn, if i am going it right, i can not make that bug that puzzles has
08:09:54puzzlescan't reproduce it?
08:10:39advcomp2019yea.. i have a sansa e280
08:10:49puzzlesmaybe it's specific to the ipod then
08:11:50 Join AndyI [0] (i=AndyI@
08:12:16Lloreanpuzzles, advcomp2019: What SVN revision is each of you using, and what audio codec was the music you played before using mpegplayer?
08:12:26 Join Acky [0] (
08:12:46advcomp2019Llorean, i just download r18896
08:13:52amiconnadvcomp2019: I can reproduce it on the H10
08:14:47advcomp2019how are you doing it.. i can try it again
08:15:06puzzlesalso seems to happen when playback is stopped
08:15:48amiconnadvcomp2019: Play some music. *pause* (not stop) music. Then go to the browser and start the video
08:16:04amiconnpuzzles: It definitely works when audio is stopped before.
08:16:36puzzlesi'll try again
08:17:20advcomp2019still audio.. unless it is because i have the video on the microsd card
08:18:01amiconnMaybe it doesn't affect the sansa.
08:18:14*amiconn needs to rebuild his c200 build though before he can test that
08:18:41puzzlesno, sound is not working when stopped
08:20:30advcomp2019puzzles, what type of music are you using before going to the video?
08:20:54puzzlesmp3 in this case
08:21:18advcomp2019o ok.. so still cant then
08:22:27advcomp2019even with music playing then going to the video, i still get audio
08:24:57 Quit jhulst (Read error: 113 (No route to host))
08:28:29 Quit bertrik (Remote closed the connection)
08:30:03 Quit Acksaw (Connection timed out)
08:31:08amiconnadvcomp2019: Music playing shouldn't trigger the bug iiuc. It's the pause that causes the problem
08:31:18 Quit bughunter2 ("bye")
08:31:50 Join ender` [0] (
08:32:10amiconnIt may be that not al targets are affected by this
08:33:24advcomp2019i think you are right have tried it about four times, and have not go it yet
08:33:47*puzzles -> bed, highlight me if you need anything more from me
08:37:43 Quit GodEater (" ajax IRC Client")
08:37:44 Quit cool_walking_ (" ajax IRC Client")
08:40:44 Quit perrikwp (" ajax IRC Client")
08:42:24 Quit BigBambi (Read error: 113 (No route to host))
08:44:13 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
08:46:00 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:46:24 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
08:47:50 Join Rob2223 [0] (
08:48:39soapthat's good to know, amiconn, last time I recall seeing the breakdown on the old version of the wikipage the G5 was a bit slower than the G4 (I think it was due to the slower screen?)
08:49:06amiconnThat must've been a looong time ago
08:50:00amiconnIt was due to the wait-for-bcm. Nowadays we're writing directly to the bcm memory without waiting, and trigger the updates asynchronously
08:50:16amiconnOh, and back then the YUV->RGB conversion was written in C
08:50:47*amiconn checks when he added the new bcm update mechanism
08:51:44amiconnAlmost a year ago...
08:56:23 Join n1s [0] (n=nils@rockbox/developer/n1s)
09:01:52 Join GodEater [0] (i=c2cbc962@gateway/web/ajax/
09:03:03 Join stoffel_ [0] (
09:03:27amiconnadvcomp2019: Sansa does indeed not suffer from this problem
09:05:10 Quit Rob2222 (Read error: 110 (Connection timed out))
09:05:17advcomp2019o ok good because i have be testing a few things and have not got it too
09:09:23amiconnipod Photo does though, as does the already mentioned H10 (small)
09:20:51 Join petur [50] (n=petur@rockbox/developer/petur)
09:37:04 Join Thundercloud [0] (
09:40:43 Join orly_owl [0] (
09:41:18orly_owlI wasn't in this channel. No wonder it was so quiet. >_>
09:44:03 Quit ajonat ()
09:44:49 Quit daurnimas (Connection timed out)
09:49:08 Quit stoffel_ (Read error: 113 (No route to host))
09:52:10 Join Nibbl [0] (
09:53:38 Join avis [0] (n=ident@pdpc/supporter/student/avis)
09:56:22 Join culture [0] (
10:00:02 Join BigBambi [0] (i=86ceaf40@rockbox/staff/BigBambi)
10:00:23 Join daurnimator [0] (
10:04:51 Join stoffel_ [0] (
10:05:30 Join DerDome [0] (
10:05:36 Quit linuxstb (Read error: 110 (Connection timed out))
10:06:26***Saving seen data "./dancer.seen"
10:08:18 Quit DerDome (Client Quit)
10:09:37 Quit daurnimator (Read error: 113 (No route to host))
10:10:46 Quit Nibbl (Read error: 113 (No route to host))
10:12:05 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
10:12:38 Join einhirn [0] (
10:12:41 Join Nibbl [0] (
10:20:58 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
10:21:47*linuxstb posts a long comment to the BMP resize patch and hopes others will also look at it and comment -
10:23:58 Quit Thundercloud (Remote closed the connection)
10:25:04B4gderlooks like very good feedback and comments
10:25:18B4gderI didn't check out the recent versions of the patch
10:31:50*jchillerup wrote some code for modifying BMP files for embedded systems literally two weeks ago
10:31:57jchillerupNo resizing though :)
10:32:51 Join kushal_12_27_200 [0] (n=kushal@
10:33:54GodEaterB4gder: have you seen the email from Stephen about repointing our A record for the Positive mirror ?
10:34:43LinusNlinuxstb: the 3-byte structs might very well be aligned, thus wasting 1 byte per item
10:34:53peturGodEater: have you not seen the reply?
10:35:50GodEaterpetur: typical, I just saw it :/
10:37:04linuxstbLinusN: Yes, but that struct isn't always used in arrays (or just in small arrays, like the bitfields array). So maybe using two different rgb types would be more efficient overall.
10:38:22B4gdercommitter #66 is now added, welcome funman!
10:40:57 Join thegeek_ [0] (
10:41:01 Quit kushal_12_27_200 ("Leaving")
10:41:05 Quit stoffel_ ("leaving")
10:49:36 Join funman [0] (
10:53:37 Join {phoenix} [0] (
10:53:49 Quit thegeek (Read error: 110 (Connection timed out))
10:54:38 Join stoffel_ [0] (
11:02:25funmanBagder: do you moderate emails sent to rockbox-committers by non subscribers (I had sent one this week-end but maybe I should send it again)
11:02:49B4gderonly subscribers are let through, yes
11:04:21funmansaratoga: did you see FFmpeg was developing a new AAC decoder to rip the dependency on libfaad?
11:05:33funmanhm I will rather answer on the development ML
11:05:49 Quit culture (Read error: 110 (Connection timed out))
11:12:48linuxstbfunman: That aac decoder has been in development for a while. IIUC, it doesn't support all the features of AAC that libfaad does, and (I think) it's also floating point.
11:13:23funmanI remember the developer was interested by a fixed point implementation (and also that Low Complexity was the primary goal)
11:16:48 Quit spiorf (Remote closed the connection)
11:17:01funman(he is superdump on irc)
11:20:19 Nick Darksair is now known as Darksair{away} (n=user@
11:20:48funmanCOMMITTERS file is in ISO8859 encoding, shouldn't it be UTF8 ?
11:21:21peturfirst commit coming up? ;)
11:21:36funmanno, no .. :o
11:22:14B4gderfunman: yes, it _should_ but it's like that because most of the web site is still 8859 and the texts from that file is used there...
11:22:29B4gderit's one of them "yeah we should fix that one day" things
11:23:08funmanwhen ႰႹႲႭႦ will become committer I guess ;)
11:23:29B4gderwe'll just ask him/her to change name first ;-P
11:26:43funmanit seems I can't dcommit from the git repository, I have to use svn first to store my authentication informations
11:30:38 Quit {phoenix} ("Konversation terminated!")
11:33:33 Join {phoenix} [0] (
11:35:21funmanI don't want to download the full SVN repository again :(
11:36:45B4gderyou should be able to svn commit from an already checked out svn repo
11:36:57B4gderbut not from a git repo of course
11:37:27 Quit linuxstb (Read error: 110 (Connection timed out))
11:37:39funman explains how to use git-svn to commit from a git repository coupled to the svn repo
11:38:43 Quit avis ("Ex-Chat")
11:42:26funmanshame on me : it was just a missing perl module
11:43:26*GodEater hands funman his coat
11:43:39GodEaterdon't let the door hit you on the way out
11:44:05funmanouch - too late
11:44:35orly_owloh snap
11:49:15 Quit Hadaka (Remote closed the connection)
11:49:31orly_owlWhere's the page about the firmware for some of the Archos players?
11:49:50orly_owlThe firmware that replaces the stock firmware.
11:50:29GodEaterwhich would be what rockbox does
11:50:57B4gderthat'd be
11:51:03orly_owli saw a /. article talking about this
11:51:33orly_owlrockbox is installed to the flash memory, not the hard drive
11:51:54B4gderorly_owl: that's not true on many devices
11:52:13orly_owlit is for some
11:52:25orly_owllike the archos ondio
11:52:29 Quit ameyer (Read error: 110 (Connection timed out))
11:52:32B4gderondio only has flash
11:52:39B4gderso it's hard to use a harddrive then
11:52:52orly_owlyeah but rockbox is loaded onto the firmware chip
11:53:00B4gdernot necessarily, no
11:53:30B4gderif you by that mean the nor flash
11:53:39orly_owlyeah i think that thing
11:54:02orly_owli read somewhere that it didnt need the original firmware anymore, even to transfer files
11:54:22B4gderrockbox doesn't need the OF on Archos
11:54:37orly_owlyeah. got a link to that?
11:55:45 Nick Darksair{away} is now known as Darksair (n=user@
11:56:03orly_owlNever mind.
11:59:10 Quit {phoenix} (Read error: 110 (Connection timed out))
12:01:33 Join {phoenix} [0] (
12:03:27funmanis it possible to know the MEMORYSIZE set in tools/configure from a .c file ?
12:04:04B4gderthat's passed on in the MEM define in most makefiles
12:06:09JdGordonis there a reason for both those defines?
12:06:18funmanah right
12:06:29***Saving seen data "./dancer.seen"
12:06:40JdGordonMEM should be defined for every target and MEMORYSIZE should be removed (or the other way around...)
12:06:56*JdGordon spotted that annoyance in the dircache code today
12:07:15B4gderas MEMORYSIZE is exported in the root makefile I don't think we need any other
12:07:25B4gderand MEMORYSIZE seems a better name to me than just MEM
12:08:08B4gderhm, but that's just a variable, not a define
12:08:15B4gderthe define is probably only MEM isn't it?
12:08:40 Quit reacocard (Read error: 60 (Operation timed out))
12:09:01 Join reacocard [0] (n=reacocar@WL-112.CINE.HMC.Edu)
12:09:07JdGordonthis sort of line is bad... #if ((defined(MEMORYSIZE) && (MEMORYSIZE > 8)) || MEM > 8)
12:09:17JdGordoneven i it has been there for years
12:13:29*funman crosses fingers hoping that svn's mkamsboot will not brick his Clip
12:17:44funmanfine :)
12:19:59 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:20:40funmanlinuxstb: do you still have a problem with home button on your clip ?
12:21:05linuxstbYes, the dualboot.S code in SVN is the same as when it was first committed - I haven't tried fixing it.
12:21:23linuxstbAlthough you gave me a patch to try once, and that didn't seem to help.
12:21:35 Quit JdGordon ("Konversation terminated!")
12:23:06 Join moos [0] (
12:23:37 Join JdGordon [0] (
12:25:36 Quit stoffel_ ("leaving")
12:28:29 Quit Dhraakellian (Read error: 110 (Connection timed out))
12:34:12 Join kushal_12_27_200 [0] (n=kushal@
12:41:11 Join faemir [0] (
12:45:10 Join goffa [0] (n=goffa@
12:46:12 Join Hadaka [0] (
12:53:00funmanlinuxstb: can you try this last commit ?
12:54:11linuxstbMy Clip is at home at the moment, and I'm not. I can try this evening though.
12:54:27funmanah right, I forgot normal people have jobs sorry ;)
12:54:52funmannow should I commit my not working code for the SD interface ?
12:55:28funmanat least I suppose it lets us communicate with the SD controller since I receive meaningful answers
12:57:35funmanalso when is the build table on updated ?
12:58:33moosfunman: that's at your assesment...
12:58:51 Quit goffa_ (Read error: 110 (Connection timed out))
12:58:51moossince that will break nothing....
12:59:09funmanthat is obvious ;)
12:59:15 Join FOAD [0] (
12:59:25moosthen go commit! :)
12:59:26BigBambifunman: It is updated on every commit that isn't just e.g. manual
13:00:38funmanI don't see it rebuilding for my last commits
13:00:49BigBambiIt might be stuck
13:01:11*BigBambi doesn't know and leaves it to someone that knows slightly more than him
13:03:17moosfunman: the build system seems to not was trigered, let's wait in B4gder to see what's hapened....
13:06:16funmanI suppose there is no red since the sansav2 targets aren't built ;)
13:06:30moosgood suposition :)
13:11:11 Quit JdGordon (Read error: 104 (Connection reset by peer))
13:12:46linuxstbfunman: Something definitely seems broken - I would have expected at least some of your commits (i.e. not anything in rbutil/mkamsboot) to trigger a rebuild. Your commits are also not on the front page.
13:13:51 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
13:14:27funmanI noticed that - but the mail went to rockbox-dev anyway
13:15:58 Quit n1s ()
13:17:22 Quit kushal_12_27_200 ("Leaving")
13:17:28 Join CaptainKewl [0] (
13:20:30 Join Seed [0] (
13:26:56 Quit pixelma2 (Nick collision from services.)
13:27:04 Join pixelma2_ [0] (n=marianne@rockbox/staff/pixelma)
13:27:06 Nick pixelma2_ is now known as pixelma2 (n=marianne@rockbox/staff/pixelma)
13:27:07 Quit amiconn (Nick collision from services.)
13:27:14 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
13:27:32B4gdersomehow my repo turned locked
13:30:49funmanmaybe the 2 pretty fast consecutive commits ?
13:31:23B4gderno, it's something stranger than that
13:31:38B4gderand most likely something just in my end
13:41:58 Join lasser [0] (
13:48:50 Quit einhirn (Connection reset by peer)
13:58:15 Join pvbcharon [0] (n=charon@
14:02:40 Quit FOAD (Remote closed the connection)
14:02:50 Join FOAD [0] (
14:06:33***Saving seen data "./dancer.seen"
14:06:36 Join robin0800 [0] (
14:10:39preglowB4gder: is curl included by default on os x or did i install some time?
14:10:54B4gderby default
14:11:52LinusNexcept it is a version from when jesus had diapers
14:12:30preglowisn't the usual bsd thing "fetch" ?
14:12:46B4gderyes it is
14:12:52preglowthat doesn't seem to be here
14:12:56preglowasked them to include a newer one? :>
14:13:06funmanosx isn't 'bsd' : it uses gnumake by default for example
14:13:17preglowfunman: yeah, but it's more bsd than anything else
14:14:06preglowthat much is obvious to anyone used to gnu tools
14:14:15preglowi still botch half my attempts at using the tools, heh
14:14:34B4gderyes, they seem to try to avoid gpl licensed stuff if they can
14:15:25 Quit robin0800 (Remote closed the connection)
14:19:21 Join J-23 [0] (
14:20:05J-23anything new to test on e200v2?
14:21:16funmanJ-23: nope, do you have something new ? ;)
14:22:00J-23funman: I'm not good in low-level embedded programming
14:22:34funmannot good is enough to look at it ;)
14:24:50J-23do you need Terribly High Resolution™ PCB scans?
14:25:38funmanI don't work on the e200 so personally I'm fine
14:25:53funmanI think daniel_at uploaded some pictures already
14:26:05 Quit {phoenix} (Remote closed the connection)
14:28:29linuxstbJ-23: There is still high-level code that needs to be written for the V2s. One job is to get the UI simulators compiling - which involves implementing things like button mappings in Rockbox itself.
14:28:34 Quit CaptainKewl (Remote closed the connection)
14:29:54funmanI'm not sure if firmware (not bootloader) builds either
14:29:55 Join evilnick [0] (i=0c140464@gateway/web/ajax/
14:31:05linuxstbfunman: No, it won't. Getting the sim building will be a large part (and the more tedious part) of that work though.
14:31:22 Join meven [0] (
14:34:28JdGordon is part of the init code for dircache... that happens even if dircache is disabled (and not doing it will crash if you try going into the file browser)... is there any reason to not just change the struct to statically allocate that instead?
14:35:16JdGordonI would have thought that if dircache is disabled all calls just become thin wrappers for dir_uncached functions, but apparently not
14:35:37B4gdera fixed amount of buffer_alloc() seems pointless to me
14:40:18 Quit funman ("leaving")
14:41:15 Join Acksaw [0] (
14:42:32 Quit pvbcharon ()
14:44:30JdGordonthe "less likely to break things" fix is to add a char entrybuf[MAX_OPEN_DIRS][MAX_PATH]; and fix the pointers to that, workable?
14:44:30 Nick Darksair is now known as Darksair{away} (n=user@
14:45:16 Quit wpyh (Read error: 104 (Connection reset by peer))
14:45:19B4gderI'd say its even preferable
14:45:30 Part B4gder
14:48:56 Part LinusN
14:55:01 Join {phoenix} [0] (
14:58:08 Quit Acky (Connection timed out)
15:00:01 Join Schmogel [0] (
15:01:45 Join pvbcharon [0] (n=charon@
15:06:34 Join LambdaCalculus37 [0] (
15:10:08 Join ameyer [0] (
15:12:46 Nick Darksair{away} is now known as Darksair (n=user@
15:13:34 Join micleftic [0] (n=54aef0c4@gateway/web/cgi-irc/
15:14:42miclefticI have no clue if any of the developers of Rockbox are here, but let me tel you: THANKS SO MUCH!" I love this project so much, it really improved my Sansa e280 a lot. Unbelievable what can be done.. so thanks a lot and keep on working!
15:15:31 Quit micleftic (Client Quit)
15:19:51 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:20:03 Quit LambdaCalculus37 ("Ka-chunka")
15:22:35 Join wpyh [0] (n=william@
15:23:10 Join stoffel_ [0] (
15:36:33 Quit Darksair ("Use the Force, Luke!")
15:53:12 Quit lasser ("ChatZilla 0.9.83 [Iceweasel 3.0.3/2008092816]")
16:02:45 Join funman [0] (
16:06:35***Saving seen data "./dancer.seen"
16:07:20 Join blahrus [0] (n=blahrus@
16:09:03 Join ze0 [0] (i=ze@
16:09:27funmanI need to use dynamic register addresses for the AS3525 SD, because the same driver can be used with the flash, and with the SD slot for targets with it
16:09:46funmanthe only difference would be the base address of all registers
16:10:10 Join Darksair [0] (n=user@
16:10:16 Quit ze (Read error: 113 (No route to host))
16:10:16 Nick ze0 is now known as ze (i=ze@
16:11:15 Quit TMM (Read error: 104 (Connection reset by peer))
16:16:14 Quit blahrus ("Ex-Chat")
16:20:40 Quit Darksair ("(define zero (lambda (f) (lambda (x) x)))")
16:24:27 Join n1s [0] (n=nils@rockbox/developer/n1s)
16:24:44 Join Darksair [0] (n=user@
16:28:58 Join blahrus [0] (n=blahrus@
16:38:00 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
16:38:36 Join dirwiz [0] (
16:39:08 Part dirwiz
16:39:18 Join dirwiz [0] (
16:41:27 Join TMM [0] (
16:44:22 Quit puzzles (Remote closed the connection)
16:44:33 Join puzzles [0] (
16:59:56 Quit evilnick (" ajax IRC Client")
17:00:46 Quit Nibbl (Read error: 60 (Operation timed out))
17:01:17 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:01:33 Quit pvbcharon ()
17:07:24 Part dirwiz
17:08:13 Quit Darksair ("updating emacs")
17:09:38 Join Darksair [0] (n=user@
17:09:41 Join cg [0] (
17:12:42 Quit perrikwp (" ajax IRC Client")
17:14:32 Join bughunter2 [0] (n=Jelle@
17:19:59J-23hmm, can I make e200v2 background image or you'll just copy it from v1 UIsim?
17:20:17 Quit meven (Remote closed the connection)
17:21:08funmanare they different?
17:21:28 Join karashata [0] (n=kimi@
17:23:50 Join kushal_12_27_200 [0] (n=kushal@
17:24:08 Join spiorf [0] (
17:24:12 Join perrikwp [0] (i=98213725@gateway/web/ajax/
17:30:28 Join Slack [0] (
17:35:32 Join meven [0] (
17:37:28J-23probably no, I didn't look thoroughly at them
17:41:53 Part {phoenix} ("Konversation terminated!")
17:52:11 Join aarcane [0] (
17:53:33 Join Nibbl [0] (
17:55:58linuxstbJ-23: You'll need to copy the v1 bitmap, and call it "UI-e200v2.bmp"
17:56:45J-23hmm, maybe symlink instead of two identifal files?
17:59:02cgi've added a simple patch for sudoku crash in the tracker, FS #9509
17:59:19cgmaybe a developer can review and commit?
18:00:56 Join robin0800 [0] (
18:02:27 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:05:07 Quit BigBambi (" ajax IRC Client")
18:05:19 Join einhirn [0] (
18:05:32 Quit einhirn (Client Quit)
18:06:08n1sJ-23: symlink is probably not so good as we have people building on windows using cygwin
18:06:36***Saving seen data "./dancer.seen"
18:07:29preglowanyone tried compiling rockbox with new gccs?
18:08:09robin0800preglow: on what platform?
18:08:17preglowos x
18:09:27preglowworks this far \o/
18:09:46robin0800preglow: I know there is a problem with linux
18:09:53preglowwhat kind?
18:10:23preglowah, there we hit the wall
18:10:28preglowin libdemac, what a surprise
18:10:29preglowregister constraint
18:10:54robin0800preglow: dosn't make properly using the script
18:14:28linuxstbWhich target?
18:15:22preglowhacking away the opts now to see if i can compile the rest
18:15:37 Join bertrik [0] (
18:16:11preglowthis is the reason why i prefer to write asm opts as full assembly functions
18:16:20preglowno damn registers constraints that mess up every new gcc version
18:16:27linuxstbJ-23: Don't worry about the e200v2 sim - I'm looking at it now, and have almost fixed the few things that needed fixing.
18:16:35preglowmore of a bother, of course...
18:16:52 Quit martian67 ("gone")
18:17:32preglowthe rest compiled just fine
18:17:34preglownow for testing...
18:17:37 Quit karashata ("I go, only to return again some time...")
18:17:42*linuxstb is happy Sandisk kept the same LCD and button layout for the V2s - makes the apps/ part of the port come for free...
18:17:45J-23anything new to test on e200v2?
18:18:15 Join kugel [0] (n=chatzill@unaffiliated/kugel)
18:19:17 Quit petur ("now sports...")
18:19:56preglowlinuxstb: anything like an "eject" command in os x?
18:19:57 Join karashata [0] (n=kimi@
18:20:28linuxstbSomething like "diskutil eject /dev/disk1"
18:20:55preglowwell, what do you know
18:21:05preglowit works
18:21:14preglowtest music
18:21:34preglowmp3 check
18:22:05 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
18:22:23preglowvorbis has problems
18:22:37linuxstbAnyone know whether I should change all lang files if I change english.lang from something like "e200: ...." to "e200,e200v2: ..." ? Or rather, is it considered lazy if I don't?
18:23:06preglowi'll forgive you if you don't...
18:23:15linuxstbThat's good enough for me.
18:23:52preglowbut i don't know if you have to, in any way
18:24:01linuxstbAnd I guess e200*: would be easier than e200,e200v2 ?
18:24:02bertrikdo we really need to distinguish between e200 and e200v in the lang files, I suppose they would use the exact same texts
18:24:21domonokyhm.. before you do all this work, maybe we should rethink the naming of those v2 ports...
18:24:22linuxstbThe lang files use model names, not keypad names
18:25:11linuxstbIf I do "e200*", then the model names won't matter...
18:25:51domonokyyes, e200* works even, when we change the names..
18:26:00preglow4.3 rockbox is some kilobytes larger
18:26:20domonokye200ams/m200ams would be good names..
18:26:54preglow4.3 generally makes code that is a wee bit bigger, it seems
18:27:06preglowhmm, no, it varies, sometimes a bit smaller, sometimes a bit larger
18:27:17bertrik4.3 what?
18:28:46 Join karashata_ [0] (n=kimi@
18:28:51bertrikI don't see how ams is a good name, it's the name of the company by the chip is called as3525
18:28:57preglowguess i should do some benchmarks
18:29:21 Quit karashata (Nick collision from services.)
18:29:28 Nick karashata_ is now known as karashata (n=kimi@
18:30:51linuxstbdomonoky: I guess I'm just used to calling them "v2" (but I know about the troublesome m200s...)
18:31:03domonokybertrik: then suggest another postfix :-) m200v2 is bad, as its essentially a m200v4..
18:32:13bertrikok I agree about the m200v2/v4
18:33:11 Join Strife89 [0] (n=michael@
18:33:18bertrikand m200as3525 would be a bit long... :)
18:33:25n1spreglow: i tested 4.3 for coldfire some time ago, it required a few changes to rockbox to build and work ok, ther's a patch in the tracker but it's probably pretty outdated by now...
18:33:37 Quit n1s ()
18:34:11cgwhoops, i accidentally entered the patch FS #9509 in the bugs section instead. maybe someone can fix? (or just apply the patch and close the task;)
18:34:34bertrikI'll fix it
18:36:10linuxstbcg: Sorry to criticise a one-line patch, but maybe a one-line comment to accompany it would be helpful? ;)
18:37:18cgcomment in the code? sure i can add that if it seems necessary
18:38:36 Quit stoffel_ (Read error: 113 (No route to host))
18:39:07 Nick karashata is now known as dalgarath (n=kimi@
18:39:10 Join robin0800_ [0] (
18:39:27 Nick dalgarath is now known as karashata (n=kimi@
18:39:58 Quit robin0800 (Read error: 104 (Connection reset by peer))
18:40:08 Join {phoenix} [0] (
18:42:08 Quit Darksair ("People who are zhuangbility want to show their niubility but only reflect their shability.")
18:44:31 Quit domonoky (Read error: 60 (Operation timed out))
18:44:36 Join Thundercloud [0] (
18:44:48cglinuxstb: done
18:45:06 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:45:57 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:46:41 Join miepchen^schlaf [0] (
18:47:09linuxstbcg: Thanks. If bluebrother (who has been looking at sudoku patches recently) doesn't do it, I'll try and commit later tonight
18:47:11 Join Darksair [0] (n=user@
18:47:15*amiconn hrmphs at the current firmware/target/arm/ structure
18:47:43linuxstbamiconn: Any idea?
18:47:50linuxstbI mean ideas? (to improve it)
18:48:48amiconnIt's currently a mix between the old (real-manufacturer) structure and the refined arm-subarch structure
18:48:57amiconnSomeone needs to perform the necessary cleanup
18:49:02 Quit blahrus ("Ex-Chat")
18:51:53amiconnipod, iriver, olympus, philips, sandisk and tatung are all PP targets (although I'd probably split PP5002 and PP502x)
18:51:56 Join mf0102 [0] (
18:52:09 Quit {phoenix} (Remote closed the connection)
18:52:33amiconnarchos is a tsm320* target iirc, but not tms320dm320
18:54:00 Quit grndslm (Remote closed the connection)
18:54:05 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
18:55:47soapThis may be the stupidest question ever, but why a e200v2 sim? What will that sim do different than the v1 one? Same resolution, same player image, same keymap...
18:55:50linuxstbYes, just cleaning up the portalplayer stuff would do most of it.
18:56:34 Join dany_21a_ [0] (
18:56:40linuxstbsoap: Just for completeness... There are some differences, depending on how accurate the sim becomes - e.g. the e200v2 only has 8MB RAM, compared to 32MB for the v1. We may discover other things.
18:56:43amiconnsoap: You can ask the same question for archos fm recorder vs. recorder v2, mini G1 vs. mini G2, ...
18:57:14 Quit bughunter2 ("bye")
19:00:07 Join Dhraakellian [0] (
19:02:58 Quit miepchen^schlaf ()
19:06:24 Join culture [0] (
19:06:29 Quit karashata ("I go, only to return again some time...")
19:06:39 Join karashata [0] (n=kimi@
19:07:17 Join miepchen^schlaf [0] (
19:07:36 Quit karashata (Client Quit)
19:07:48 Join karashata [0] (n=kimi@
19:09:56 Quit perrikwp (" ajax IRC Client")
19:10:37 Quit robin0800_ (Remote closed the connection)
19:18:48 Quit Darksair ("Zzz...")
19:19:04 Join pvbcharon [0] (
19:20:30 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:20:50 Quit pvbcharon (Client Quit)
19:21:05Strife89I take it the Sansa v2 work is coming along well?
19:21:38funmanif you have some code to share, sure :)
19:21:48*Strife89 has been reading the commit list.
19:21:58Strife89funman: Sorry, no code on me. :(
19:22:17Strife89No v2s either.
19:22:38funmanfirst steps have been made at least
19:23:32 Quit kushal_12_27_200 ("Leaving")
19:23:40 Join kimi [0] (n=kimi@
19:24:01 Quit karashata (Read error: 113 (No route to host))
19:24:19 Nick kimi is now known as karashata (n=kimi@
19:24:43 Nick karashata is now known as kimi (n=kimi@
19:25:48 Nick kimi is now known as karashata (n=kimi@
19:26:09 Nick karashata is now known as kimi (n=kimi@
19:27:32 Quit linuxstb (Read error: 110 (Connection timed out))
19:29:05 Nick cg is now known as cg|aw (
19:29:10 Nick cg|aw is now known as cg (
19:30:13 Join stoffel_ [0] (
19:31:12 Join karashata [0] (n=kimi@
19:31:29 Quit kimi (Read error: 113 (No route to host))
19:31:55 Nick karashata is now known as kimi (n=kimi@
19:32:18 Nick fxb__ is now known as fxb (
19:33:22 Quit kimi (Client Quit)
19:33:27 Join kimi [0] (n=kimi@
19:34:27 Nick kimi is now known as karashata (n=kimi@
19:36:11 Join dirwiz [0] (
19:37:24 Join dany_21a [0] (
19:38:48 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:39:08 Join blahrus [0] (n=blahrus@
19:41:05 Join Lear [0] (i=chatzill@rockbox/developer/lear)
19:42:19LearSo, linuxstb has sneaked away? :)
19:45:16 Join bughunter2 [0] (n=Jelle@
19:46:03 Quit Horscht ("")
19:47:49 Quit dany_21a_ (Read error: 110 (Connection timed out))
19:51:04 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:55:44 Join neddy [0] (n=john@nat/sun/x-68d8cd56acebc738)
19:58:36 Nick karashata is now known as dalgarath (n=kimi@
19:58:42 Nick dalgarath is now known as karashata (n=kimi@
20:06:38***Saving seen data "./dancer.seen"
20:07:27LearBagder: around? Linuxstb's last commit exposed what looks like a bug in genlang...
20:09:48 Quit bertrik (Remote closed the connection)
20:11:15 Quit jhulst (Read error: 113 (No route to host))
20:12:34 Join bertrik [0] (
20:13:56 Quit gevaerts (Nick collision from services.)
20:14:05 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
20:14:32 Part Dhraakellian ("Meddle not in the affairs of ircops, for they are (not so) subtle, and quick to anger.")
20:17:13denes_could anybody tell me where I can find the infamous sansa v2 git tree?
20:17:43 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
20:17:45denes_s/infamous/famous sorry :)
20:17:54denes_english is not my native language
20:18:09 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
20:18:41*linuxstb goes to the corner and checks his commit
20:18:43Learlinuxstb: familiar with genlang by any chance?
20:19:07linuxstbNo... Did I do something obvious?
20:19:08bertrikdenes_, AFAIK all important stuff from sansa v2 git is now in SVN
20:19:22 Join Schmogel [0] (
20:19:28linuxstb(apart from not test-compiling)
20:19:29Learlinuxstb: Reason the build fails: genlang translates "e200" into the regex "e200..*".
20:19:45Learsorry, "e200*" into...
20:19:46denes_bertrik: okay. and the git tree doesn't have a web interface?
20:19:57linuxstbWhy would it do that?
20:20:00denes_domonoky: thanks
20:20:25domonokyi think the only thing not in svn is the not-working sd/nand code
20:20:29 Quit Thundercloud (Remote closed the connection)
20:20:47amiconnlinuxstb: My SH1 asm version of nrv2e_d8.S is 45% faster than the C version and less than half the size :)
20:20:58linuxstbNice ;)
20:21:08Learlinuxstb: Not sure, really. First it replaces "*" with ".?*", which isn't likely to work. So to fix that (or something completely different) it then replaces "?" with "."...
20:21:53LearMaybe the "?" replacement is to simulate simple "*" and "?" patterns.
20:21:55 Join dabujo [0] (
20:22:48LearMaybe removing the "?" in the first replacement is the right fix...
20:24:05linuxstbUnless of course that was done for a reason...
20:24:10LearSeems to have the intended effect here.
20:24:53LearDoubt that; ".?*" isn't a valid pattern according to the perl version I have here. :)
20:25:00 Part dirwiz
20:25:16 Quit karashata ("I go, only to return again some time...")
20:28:51 Quit pixelma2 ("-")
20:29:05 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
20:29:10 Join goffa_ [0] (n=goffa@
20:29:44 Quit moos ("Rockbox rules the DAP world")
20:30:55 Join pixelma_ [50] (i=pixelma@rockbox/staff/pixelma)
20:31:17 Quit pixelma (Nick collision from services.)
20:31:18 Nick pixelma_ is now known as pixelma (i=pixelma@rockbox/staff/pixelma)
20:31:47funmandany_21a: can you make the link between the OF and the e200v2 lcd code in svn ?
20:33:01linuxstbLear: No, but ..* is, but that would be better written as ".+" if it was intentional. Have you test-compiled many targets?
20:33:40LearProblem with "..*" is that "e200" doesn't match "e200*", which it should.
20:33:56LearHaven't tested many targets yet... Just e200 so far.
20:34:08linuxstbfunman: Part of sim commit was to fix the target directory (t_model) for the e200v2 - it was using sansa-clip before. I don't know what that affected at this stage though (re: backlight/lcd testing)
20:35:21linuxstbLear: Looking at the lang file, I couldn't see anything that would break with your change, so I think it's OK.
20:35:40LearI'll test a few builds here first.
20:36:09funmanlinuxstb: ah right ..
20:36:23*funman calls for e200v2 testers: just build & run what is in svn
20:37:09funmanmaybe the regexp was ipod* for "ipod-1g" and "ipod-2g" ?
20:37:15funmani.e. it assumed a suffix
20:37:56 Join perrikwp [0] (i=982131d9@gateway/web/ajax/
20:38:25 Join Dhraakellian [0] (
20:39:15Dhraakellianhmm... so that huge mp3 optimization didn't actually make it into 3.0?
20:40:02 Quit goffa (Read error: 110 (Connection timed out))
20:40:25bertrikDhraakellian, you mean the dual-core thing for the mp3 codec?
20:40:39linuxstbDhraakellian: No, but there's no real reason not to use a current build if that feature is useful to you.
20:40:58Dhraakellianheh... it's not really terribly useful to me, since I use Vorbis for most stuff
20:41:06Dhraakellianbut I was just thinking of doing a battery test
20:41:16*domonoky reads about this low-mem target discussion, and notices the lowest mem we have is 2MB with greyscale display. Anything with color has atleast 8MB..
20:42:04Dhraakellianwhich was keeping me from updating, since I wanted to test with 3.0
20:43:40Dhraakellianis it sad that Sudoku improvements are more of an incentive for me to upgrade my music player's firmware than improvements in actual music playback time?
20:44:06BigBambiyes :)
20:45:13Dhraakellianjust out of curiosity, is it at all likely that other codecs could see such a drastic performance boost, or was this something very mp3-specific?
20:45:57funmanIf I understand correctly every codec has to be modified to execute in parallel
20:46:13funmanalso I'm not sure what are the capabilities of the 2nd core used here
20:46:58 Join PaulJam [0] (
20:47:14Dhraakellianso if possible it's mainly just a matter of someone with the skills, time, and motivation getting around to doing it?
20:47:28BigBambipretty much
20:47:38linuxstbfunman: Yes, the codecs are all single-threaded (apart from spc, and now mp3), meaning they need adapting for dual-core.
20:47:41domonokymost codecs would probably benfit from running on both cores. but its a lot of work..
20:47:43funmanI would say it's a matter of motivating saratoga
20:47:48BigBambiIt also depends on the design of the codec if it can be easily and equally split onto dual-cores
20:48:21linuxstbdomonoky: The Clip has three colours - yellow, blue and black ;)
20:48:28BigBambi(easily being relative :))
20:48:34 Join DaaT [0] (
20:48:40*Dhraakellian wonders how much of a bounty would be needed to make Ogg Vorbis the fastest codec on the e200 again
20:48:44pixelmadomonoky: 2MB with a greyscale display - which one is that?
20:49:01funmanDhraakellian: usually multimedia coders don't work for bounties
20:49:05domonokylinuxstb: not really.. at least for the software :-)
20:49:16funmanpixelma: archos?
20:49:31pixelmathat's monochrome not greyscale
20:49:45funmangreyscale with only 2 steps ? :)
20:49:50Dhraakellianno insult intended if it at all came across as such
20:50:07domonokypixelma: it seems its even better.. all low-mem targets are b/w for now
20:50:16funmanI wonder if a better cooperation with ffmpeg is possible
20:50:18DaaThi everyone. Quick question. A couple of weeks ago I bought a used (but in great shape) iPod 5th gen, 30GB, to put Rockbox on it. First run, the battery lasted me about 8 hours, which was fine. But it's been getting worse and worse. Last night, after it was empty, I fully charged it, and today used it for 2 hours (commute), it's empty again
20:50:25DaaTanyone else this happened to?
20:50:36DaaTI listen mostly to 320kbps MP3 files
20:50:40linuxstbdomonoky: I would expect that that's not a co-incidence...
20:50:44funmanpixelma: I thought greyscale was a hack around a monochrome display - or at least I was told so
20:50:55DaaTi added all my music and just let it play till I get to work (or home), shuffled
20:51:34funmanDaaT: did you try benchmarking with the Apple firmware ?
20:51:52linuxstbfunman: We have some targets with 2-bit greyscale displays. On both mono and 2bpp displays, the "greyscale lib" uses temporal dithering to display more levels (only in plugins - it's a resource-intensive feature)
20:51:55DaaTfunman, nope
20:51:56domonokyfunman: rockbox have such a hack, all b/w displays can have 33 shades in rb.. but there are players with nativ greyscale
20:52:05DaaTis it included with their firmware?
20:52:20pixelmadomonoky: 129 shades, you're outdated ;)
20:52:29domonokyoh.. time goes by..
20:52:36funmanDaaT: I mean just use it and see how much time it takes to go out of battery
20:52:41DaaTahh :)
20:53:19DaaTdidn't like I said. Didn't even bother much with it in "apple-mode", since I only bought it for rb. Will have to put some music in it and give it ago
20:53:31Dhraakellianwhat's the best way to test OF battery life?
20:53:34 Join n1s [0] (n=nils@rockbox/developer/n1s)
20:53:53funmanthe file is grEyscale.c but its content use grAyscale : i'm not sure what is the correct wording ..
20:53:58Dhraakellianplug it into the soundcard, then hit play and record in Audacity at the same time?
20:54:07gevaertsThat's a good way, yes
20:55:19linuxstbLear: Thanks for fixing (I hope) genlang.
20:57:08pixelmawill ondio* still match ondiofm and ondiosp etc.?
20:58:13linuxstbLear's change means that "*" now means "zero or more characters", instead of "one or more"
20:59:20 Quit funman ("leaving")
20:59:34pixelmathanks for explaining - only very very little knowledge about regex here
20:59:54 Quit stoffel_ ("leaving")
21:04:28 Join IudeX [0] (n=52a0f8f7@gateway/web/cgi-irc/
21:04:33IudeXhi all
21:04:47IudeXAre you talking about Clip LCD?
21:05:28IudeXWoW it's impossible -> Funman is not here :D
21:05:35n1sdoesn't the clip have an OLED display? ;P
21:06:05 Join dabujo_ [0] (
21:06:30IudeXOch.. yeah
21:07:24IudeXSo how will rockbock looks on Clip :D <- u understand?
21:07:38 Join spiorf_ [0] (
21:09:53 Quit perrikwp (" ajax IRC Client")
21:11:00pixelmaIudeX: like this ;)
21:11:01bertrikIudeX, a screenshot was posted for clip and m200 on the sansa v2 forum thread IIRC
21:11:16 Quit dabujo (Read error: 60 (Operation timed out))
21:11:58IudeXWoW nice :) gz
21:12:07n1scg: in your patch, shouldn't the size be a sizeof(something) instead of a magic number?
21:12:09pixelmajust a quick mockup I did last time the Clip's LCD was discussed
21:13:04IudeXbut rockbox doesn't work :(
21:13:26n1swould for example statusbar+list title fit in the yellow part?
21:14:43pixelmaI've been told that this part was 16 pixels tall, so with sysfont (or another with the same height) it should
21:14:52cgn1s: it is just a copy paste row from another line, so i did not change it
21:15:34n1spixelma: then that could work out pretty nicely
21:15:38n1scg: ah, ok
21:15:47 Quit miepchen^schlaf ()
21:17:08 Quit IudeX ("CGI:IRC")
21:17:24 Join miepchen^schlaf [0] (
21:20:24 Quit Strife89 ("Bye, guys.")
21:21:56 Quit spiorf (Read error: 110 (Connection timed out))
21:21:59 Join merbanan [0] (n=banan@
21:24:54 Quit blahrus ("Ex-Chat")
21:31:04 Quit J-23 (Read error: 110 (Connection timed out))
21:37:20bertrikI think I have an explanation for why a read is done at a specific SDRAM address during SDRAM initialisation (required for the clip for example)
21:38:31 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
21:40:05*linuxstb waits...
21:41:20bertrikit sets the SDRAM mode register, see also
21:43:05bertrikthe value 0x2300 sets it to non-burst mode, CAS latency 3, sequential burst ordering, burst length 1 word
21:43:07 Join funman [0] (
21:43:29bertrikoh hi funman, I was just explaining the weird SDRAM read during SDRAM initialisation
21:43:32funmanbertrik: how can such a register be set only with a read ?
21:43:44funmanyeah I just read the logs ^^
21:44:26bertrikthe settings are encoded in the address for this command
21:45:48funmanah right
21:45:58funmando you know how to decode the address?
21:46:29funman0x2300 = 10001100000000b , a bit more than 10 bits
21:46:31 Part dany_21a
21:47:58domonokyah, the intend is not to read something, but to pull some address lines high :-) makes sense..
21:48:53 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
21:49:34 Quit spiorf_ (Remote closed the connection)
21:49:37bertrikI looked at the SDRAM wiki and made an educated guess that it should be shifted :P
21:50:54funmanfor the e200v2 it's 0x2300*4, or 0x2300<<2
21:51:01 Join dany_21a [0] (
21:51:02amiconnTypical SDRAM setup method - the coldfire irivers also do this
21:51:24funmanmaybe the addresses aren't encoded the same way on the e200v2 and clip/m200v2 SDRAM
21:51:59 Join perrikwp [0] (i=d1a8d351@gateway/web/ajax/
21:55:32funmanamiconn: in which file?
21:57:30 Join Rudy77 [0] (
21:59:52 Quit dabujo_ (Read error: 104 (Connection reset by peer))
22:01:18 Quit Nibbl (Read error: 113 (No route to host))
22:06:40***Saving seen data "./dancer.seen"
22:07:17 Quit PaulJam (".")
22:09:00 Quit Acksaw (Read error: 104 (Connection reset by peer))
22:09:20 Join Acksaw [0] (
22:09:46linuxstbfunman: I've just tested the svn mkamsboot on my Clip, and it's working perfectly.
22:09:51funmanbase+0x2324 <= was this calculated, or is it using the same value than the OF ?
22:09:55funmanlinuxstb: ah cool ;)
22:10:18 Quit Lear ("ChatZilla 0.9.83 [Firefox 3.1b2pre/20081024032143]")
22:11:17 Quit aarcane ("Leaving")
22:11:43funmanamiconn: at line 149 of crt0.S
22:13:11amiconnAfaik it uses the same value as the OF, but the bits are documented as well (in the SDRAM datasheet)
22:13:39amiconnLinusN should know - he did this stuff
22:13:42funmanah I think I have the datasheet for the e200v2 SDRAM (thanks to dany_21a)
22:15:16funmanrow/column are multiplexed on the same address pins
22:15:24*rasher has e200v2 sims up now
22:15:29funmanthat must mean the setting uses either column either address (or both?)
22:16:47 Quit Llorean (Connection reset by peer)
22:17:05funmanfor the e200v2 SDRAM: row length = 12, column length = 9, and we need 10 bits, so that'd be the row
22:17:07 Quit merbanan (Remote closed the connection)
22:17:24 Quit bmbl ("Woah!")
22:17:25 Join petur [50] (n=petur@rockbox/developer/petur)
22:17:31linuxstbamiconn: How big is your sh-1 ucl decompressor function?
22:17:42amiconnfunman: It's neither, because that data goes to the sdram controller, not to the sdram
22:17:50amiconnlinuxstb: 220 bytes
22:18:05amiconnThe C version is 524 bytes when not inlined
22:18:29linuxstbAre sh-1 instructions a fixed size, or variable?
22:18:36amiconnFixed, 16 bit
22:19:48linuxstbSo is sh-1 lacking some things that arm-thumb can do?
22:20:08amiconnSH1 is risc to-the-max is most respects: fixed instruction width, no shift-by-n (i.e. no barrel shifter), no condition code register (just a single bit)
22:20:57amiconnYes - SH1 never sets the T bit in arithmetic instructions, so you have to test or compare explicitly
22:21:25 Quit meven (Remote closed the connection)
22:22:28amiconnHow big was the thumb version again?
22:22:36 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:23:10 Join meven [0] (
22:23:23 Quit meven (Remote closed the connection)
22:23:41linuxstbThe version we use is 168 bytes, but I think that's got a few instructions removed from the original.
22:24:28 Join Llorean [0] (
22:24:32amiconnAh, yes
22:24:41amiconnThe plain version is 186 bytes
22:24:45 Quit Schmogel (Read error: 110 (Connection timed out))
22:25:03amiconnBut e.g. arm has load/store multiple regs - sh doesn't
22:26:43*amiconn tried the thum version on minig2 while hunting for the bug in the SH1 code
22:29:30amiconnMy final version is a bit different from the ones in UPX - it takes the same parameters as our C version, and also has the same return value
22:30:13 Join massiveH [0] (
22:30:45funmandid you write it from scratch or from gcc output ?
22:30:53amiconnThe ones in UPX take 4 parameters, and return an "error" value that indicates a decoding error when it's non-zero. Our version only takes 3 parameters and returns the compressed length
22:32:03amiconnfunman: I wrote it as a "translation" from the thumb code, with the necessary adjustments
22:33:31 Quit jhMikeS (Read error: 110 (Connection timed out))
22:34:04 Quit Horscht ("electromagnetic radiation from satellite debris")
22:35:03amiconnI'll integrate the asm version into the flash bootloader as well, for faster booting of flashed rockbox
22:35:28 Join igcbg [0] (n=4b203e89@gateway/web/cgi-irc/
22:35:39amiconnRight now only the recv1 profits from it (other archos targets don't need the self-extractor yet)
22:36:27igcbghey... new to rockbox need help downloading it for my ipod nano if u can help me out pm me please =]
22:37:15funmanigcbg: if you have a 3rd or more generation nano, rockbox will not work on it
22:37:24igcbgits a first gen
22:37:26Lloreanfunman: 2nd or more generation.
22:37:50funmanLlorean: ok, sorry
22:38:11 Join kugel_ [0] (
22:38:35Lloreanigcbg: What step in the manual are you stuck at?
22:39:18igcbgi need either the bootloader or firmware...
22:40:13LloreanNo, you need both. Why not try the automated install in the manual instead?
22:40:29 Quit balou_ (
22:40:29 Quit Rudy77 (
22:40:29 Quit goffa_ (
22:40:29 Quit cg (
22:40:29 Quit ameyer (
22:40:29 Quit FOAD (
22:40:29 Quit DataGhost (
22:40:29 Quit synergist (
22:40:29 Quit homielowe (
22:40:29 Quit Kopfgeldjaeger (
22:40:29 Quit lostlogic (
22:40:29 Quit maraz (
22:40:29 Quit liiwi (
22:40:29 Quit suom1 (
22:40:29 Quit freqmod_qu (
22:40:29 Quit gevaerts (
22:40:29 Quit neddy (
22:40:29 Quit BigBambi (
22:40:29 Quit kugel (
22:40:29 Quit Zambezi (
22:40:29 Quit BlakeJohnson861 (
22:40:29 Quit gromit` (
22:40:29 Quit rvvs89 (
22:40:29 Quit lastebil (
22:40:29 Quit tarbo (
22:40:29 Quit preglow (
22:40:38NJoinRudy77 [0] (
22:40:38NJoingoffa_ [0] (n=goffa@
22:40:38NJoincg [0] (
22:40:38NJoinameyer [0] (
22:40:38NJoinFOAD [0] (
22:40:38NJoinDataGhost [0] (n=dataghos@unaffiliated/dataghost)
22:40:38NJoinsynergist [0] (
22:40:38 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
22:40:38NJoinKopfgeldjaeger [0] (
22:40:38NJoinbalou_ [0] (
22:40:38NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
22:40:38NJoinmaraz [0] (
22:40:38NJoinliiwi [0] (
22:40:45 Nick kugel_ is now known as kugel (
22:40:55 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
22:41:13NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
22:41:26NJointarbo [0] (n=me@unaffiliated/tarbo)
22:41:53NJoinsuom1 [0] (
22:41:59 Join rvvs89 [0] (
22:43:13 Join gartral [0] (
22:43:21gartralhi sll
22:44:25NJoingromit` [0] (
22:45:48 Quit rvvs89 (Killed by (Nick collision))
22:45:56NJoinneddy [0] (n=john@nat/sun/x-68d8cd56acebc738)
22:45:56NJoinBigBambi [0] (n=Alex@rockbox/staff/BigBambi)
22:45:56NJoinZambezi [0] (
22:45:56NJoinBlakeJohnson861 [0] (
22:45:56NJoinrvvs89 [0] (
22:45:56NJoinlastebil [0] (
22:45:56NJoinfreqmod_qu [0] (i=quassel@2001:700:300:1430:213:d3ff:fee9:5ed0)
22:45:56NJoinpreglow [0] (i=thomj@rockbox/developer/preglow)
22:46:00 Join rvvs89_ [0] (
22:46:08 Quit freqmod_qu (Excess Flood)
22:46:28 Quit preglow (Read error: 104 (Connection reset by peer))
22:46:39 Join freqmod_qu [0] (i=quassel@2001:700:300:1430:213:d3ff:fee9:5ed0)
22:46:41 Quit Zambezi (SendQ exceeded)
22:47:04 Join Zambezi [0] (
22:47:11 Quit neddy (SendQ exceeded)
22:48:38 Join yhuang [0] (
22:49:27 Join lastebil_ [0] (
22:49:57 Join preglow [0] (
22:50:23 Quit igcbg ("CGI:IRC")
22:52:25 Quit mf0102 ("Ex-Chat")
22:54:19 Quit lastebil (No route to host)
22:55:01 Join Zarggg [0] (
22:56:58 Quit rvvs89 (Network is unreachable)
22:59:31Rudy77Can anyone tell me if Rockbox supports wma files below 32kbps on an IRiver H320?
23:00:36 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:00:47Bagdercan't you just try it?
23:00:55n1sRudy77: I think so but i'd guess saratoga could telll you :)
23:01:07linuxstbBagder: Did you see Lear's genlang commit?
23:01:17Bagderyes, I'm fine with it
23:02:28 Join krazykit [0] (
23:03:11Rudy77I thot I saw in a FAQ somewhere that it didn't, but I wasn't sure how old that info was. I haven't installed it yet on my player cuz of that limitation.
23:03:44saratogawhat faq
23:03:51 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
23:04:12bluebrotherisn't wma the real limitation? ;-)
23:04:53Rudy77The General FAQ at
23:04:56Rudy77It says:
23:04:57Rudy77Yes. Support for non-DRM WMA files is included, although it requires optimisation to play realtime on most targets, and will not currently play files with a bitrate less than 32kbps. See here for details: SoundCodecWMA. It is very unlikely that Rockbox will be able to play protected WMA files.
23:05:38bluebrotherRudy77: why are you reading the FAQ to us?
23:05:42n1sRudy77: if you don't like rockbox you can uninstall it
23:06:33Lloreanbluebrother: Saratoga asked which FAQ, though
23:06:45 Part gartral
23:06:55Rudy77I'm just asking if perhaps there's been an upgrade and that info is no longer accurate. I haven't installed it and don't want to go thru the process if that limitation exists right now.
23:07:12bluebrotherLlorean: well, does that mean we couldn't read the FAQ ourselves if we know which one? ;-)
23:07:12massiveHby installing rockbox, you have lost the ability to play protected WMAs now.
23:07:17Rudy77 is the FAQ.
23:07:44linuxstbYou can always test your actual WMA files in the Rockbox UI simulator -
23:08:26saratogaok i updated the faq
23:08:48saratogaand anywya if you clicked the more information link, it had up to date information
23:09:40Rudy77linuxstb: Thanks for that link!
23:12:16Rudy77saratoga: I'm sorry for my ignorance - tnx for the change - am I to conclude that it does support bitrates below 32kbps now?
23:13:10n1sWe have too many FAQ's and guides strewn about the wiki... Info regarding Rockbox and it's use should be in the manual.
23:13:51*n1s hints to people with possible codec quirks to add them to the "Supported audio formats" table in the manual (appendix A)
23:14:09Rudy77nls: I did peruse the manual and didn't see that limitation mentioned, but it was on that one FAQ, thus my confusion and it's why I asked.
23:14:47n1sRudy77: I understand your confusion, that is the problem with having similar info in many different places
23:14:55 Quit domonoky (Read error: 104 (Connection reset by peer))
23:16:39yhuangdoes pictureflow check .rockbox/albumart ?
23:19:07*rasher thought PF used the code from apps to search for albumart
23:19:14rasherBut now I'm not so sure
23:20:28yhuangi have a load of album art in .rockbox/albumart and they show up when i play a song, but when i run pictureflow it says no album art detected
23:20:46yhuangi tryed re initializing the database too
23:21:06 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
23:21:41 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
23:22:57rasheryhuang: that sounds like a bug
23:23:48bertrikI can't make pf work on my e200 either, although I think I met all requirement mentioned on the PluginPictureFlow wiki page
23:23:53rasherAs far as I can tell, it gets this info from the DB, so it should work
23:24:49 Quit massiveH ("Leaving")
23:25:54yhuangbertrik, does the album art load up in wps for you?
23:25:54 Join DerDome [0] (
23:26:43bertrikyhuang, yes. I have 120x120 AA by the way while pf prefers 100x100
23:27:58yhuangok so we know its pictureflow's problem
23:28:07funmansaratoga: do you think you will plug your jtag adaptor to your fuze before I receive the adapter I ordered?
23:29:44saratogafunman: i really have no idea when i'm going to have time to dig into this
23:30:00saratogai'm really busy this week, and then i have to figure out a way to solder to the tiny connectors
23:30:19saratogayou're likely to be done before me I think since you have normal sized leads on teh clip
23:30:28funmanI can't make you less busy, but did you investigate the suggestions on the forum ?
23:30:37saratogathe wire on a tip?
23:31:04funmancan't remember the exact thing - the only electronic I know is at an higher scale than that ;)
23:31:23pixelmapictureflow seems to have quite a number of not-so-small issues
23:31:43saratogathe two ideas i saw were to put a fine gauge wire on a soldering iron, and to use an adapter with the proper solder pitch
23:31:49rasheryhuang: I think you should file a bug
23:32:26saratogai tried the wire under a stereo microscope at 20x mag, and it was small enough to work, but I couldn't keep my hands steady enough to solder
23:32:38funmantoo much coffee ? ;)
23:33:06saratogai'll probably take another shot at it
23:33:08funmantry looking for help from other people, maybe 1 person out of x can be this steady
23:33:28scorche|shwhat pitch is this that you are trying to solder?
23:33:34saratoga0.5 mm
23:33:36bertrikis it possible to get a flatcable with the correct pitch (and use lots of flux) ? in that case you may not need to be that steady and accurate
23:34:00scorche|shi can do that if you really need
23:36:12saratogascorche|sh: whats your method?
23:36:37scorche|shnice soldering station, careful hand/eye
23:36:50saratogai should try again, it seems like i should be able to do this
23:37:56funmanread the dark tower from Stephen King, it learns how to synchronize hand and eye (to shoot, but this should work for solder)
23:38:10saratogawhich sansa is this:
23:38:36scorche| said
23:38:49saratogaah missed that
23:38:57scorche|shboth in the description and the "Products" bit
23:39:01saratogamight be easier to just buy one of those and jtag it
23:39:14saratogathey're AMS right?
23:39:18rasherThat sure was cheap
23:39:21rasherSold out..
23:39:24funmanthe v4, right
23:39:31scorche|shthe others are TCC
23:39:44funmanyou should ask domonoky or bluebrother if there is a suspect JTAG connection
23:40:50rasherBagder: thoughts on providing windows sims?
23:40:53bertrikhmm, too bad, would like one, even if it were pink :P
23:41:20Bagderrasher: I think it would be cool
23:42:12saratogaBagder: can anyone update the front page via SVN or does it take something more then that?
23:42:20Bagderrasher: if so, do you want to host them "officially" or would you like me to rsync them over somewhere?
23:42:32*linuxstb wonders if we should rename the "archos" directory in the sim, or keep it for nostalgic reasons
23:42:38Bagdersaratoga: the web site content is in svn, but the actual update from svn on the site needs to be done manually
23:43:28rasherlinuxstb: I've said many times that "disk" would make more sense, but amiconn and Bagder disagreed. Grumpy old men
23:43:39pixelmalinuxstb: is it timt for this discussion again? ;)
23:43:48pixelmatime too
23:43:52*bluebrother is nostalgic and a grumpy old man :)
23:43:53gevaertsrasher: "disk" is clearly wrong :)
23:43:55Bagdera good subject never dies! ;-)
23:44:13*linuxstb apologises - he must have missed the previous discussions (or is simply going senile...)
23:44:23rasherIt's been very brief, probably
23:44:29*gevaerts thinks that the sim should support multivolume
23:44:54pixelmaI believe that came up more than once, not sure though
23:46:06*linuxstb takes that as a "no, let's keep it as it is"
23:46:22BigBambiIf they are going to be offered more widely though, it is confusing
23:47:04funmangevaerts: if the simulator supports multiple volumes, why not 'volume1' ?
23:47:13 Quit petur ("Zzzzz")
23:47:32n1sor put_music_here ?
23:48:05n1sor s/here/inside/ even
23:48:21bluebrothernah. Why not "root"? ;-)
23:48:22Llorean"simulator data"
23:48:23pixelmawhat if people want to try out gameboy roms etc. there? ;)
23:48:37Bagderrasher: so maybe we should start with getting a script committed (somewhere) that builds all sims (cross-compiled) ?
23:48:42BigBambi"My Music" :)
23:48:54bluebrother"My Rockbox" :P
23:48:58BigBambihehe :)
23:49:00*gevaerts hits BigBambi over the head with a guitar
23:49:03Bagder"My Archos" ! ;-P
23:49:06*Bagder ducks
23:49:33bluebrotherhey, I even kinda like "My Rockbox". The only thing I dislike is this stupid space in the folder name ...
23:49:34*BigBambi was far too late in ducking
23:50:00gevaertsI actually like "root"
23:50:14rasherBagder: what kind of scripts do we have now to build stuff? Building a cross-compiled sim is exactly the same as a regular sim, except you set PATH to include the cross-compiled SDL.
23:50:24bluebrother"unzip to the root folder"? ;-)
23:50:43rasherMaybe could include a "build SDL for cross compiling the sim" option
23:50:58Bagderrasher: we have scripts like tools/release that builds a set of bins for release
23:52:21 Part Rudy77
23:52:43rasherBagder: does the main server still build all the daily bilds?
23:53:07Bagderand voices, and manuals
23:53:46rasherAny particular reason the build system isn't used for this?
23:54:15rasherExcept the voices and manuals, of course.. that's not quite as easy to distribute
23:54:50Bagderjust for the stability of knowing that it is the exact same machine doing them all, and for easyness as sometimes builds fail or there's a mixed situation exactly when the daily build is about to happen etc
23:54:56LloreanWell, it does mean that even if one build server's doing something funny, the dailies are "good" (assuming the 'funny' buildserver isn't the main one)
23:55:06rasherThat makes sense
23:57:05 Quit bmbl ("Woah!")
23:57:29*bluebrother wonders if FS will ever get fixed
23:58:13 Nick HBK- is now known as HBK (

Previous day | Next day