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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2010-10-11

00:02:51pixelmaI wanted to test a Player sim for some other reason and the build failed in some _mp3* things. Still having the patch applied so suspected it. Now that I just reverted it, the sim seems to compile... yep
00:03:48wodzso sim doesn't work or doesn't compile with my patch?
00:04:15pixelmadoesn't compile
00:05:20wodzok I'll look at it later
00:08:11wodzpixelma: did you tested recording with my patch?
00:11:57pixelmaunfortunately not yet, I suffered the crash bug with my theme on the Ondio which was fixed today and which took quite some time testing. It's resolved now and I'll try your patch next, now that my test target is working a bit better again :)
00:11:58 Quit Luca_S (Quit: CGI:IRC (EOF))
00:12:14 Quit simonrvn (Read error: Connection reset by peer)
00:12:33 Quit hebz0rl (Quit: Leaving)
00:15:48 Quit robin0800 (Remote host closed the connection)
00:17:30 Join earcar [0] (
00:17:55 Quit earcar (Client Quit)
00:22:53hondzapamaury: comment added
00:24:59gevaertswodz: it fixes plugin loading on portalplayer (and possibly other SoCs with similar caching behaviour, not sure if we have any) for plugins that have bss variables with no-cache attribute
00:25:20gevaertsWhether or not that maps exactly to greylib plugins, I don't know :)
00:26:23 Quit Rob2223 (Read error: Connection reset by peer)
00:26:29pixelmagevaerts: the Player's default WPS doesn't seem to have a progressbar. I'm preparing screenshots (from sims) to show what I mean - the Player's %pb is broken too for that matter
00:26:59 Join Rob2222 [0] (
00:27:22pixelmathat one can be shown with the shipped progressive.wps - or better by the gap in the middle of the second line which shouldn't be a gap actually...
00:28:39 Quit wodz (Quit: Leaving)
00:31:57gevaertspixelma: ok, I see progressive.wps behaving differently in a current sim than in 3.6
00:33:22pixelmaso you don't need the screenshots anymore?
00:33:38gevaertsprobably not
00:33:44pixelmaguess I'll post them anyway
00:36:22 Join Rob2223 [0] (
00:37:06 Quit BHSPitMonkey (Quit: Ex-Chat)
00:39:18 Quit Rob2222 (Ping timeout: 240 seconds)
00:41:05 Join stripwax__ [0] (
00:42:16 Quit stripwax (Ping timeout: 245 seconds)
00:44:07 Join simonrvn [0] (
00:45:12 Quit Staphylo (Quit: Bye les gens =))
00:49:41 Quit Judas_PhD (Quit: This is a quitting message)
00:50:56 Join Judas_PhD [0] (
00:56:02 Quit ender` (Quit: Only thing made by Apple that wouldn't suck would be an iVacuum.)
00:58:27 Join JdGordon| [0] (
00:58:27 Quit JdGordon| (Changing host)
00:58:27 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
00:59:05 Join krabador [0] (
01:04:25***Saving seen data "./dancer.seen"
01:12:26pixelmaJdGordon: can't seem to get the subline patch to apply (it's complaining about unexpected end again)
01:14:02 Quit Judas_PhD (Quit: This is a quitting message)
01:14:26JdGordon|meh bugger
01:15:23CIA-7New commit by 03kugel (r28239): Two new lcd/multi screen api convinience functions: draw_viewport(), fill_viewport(). ...
01:17:18CIA-7r28239 build result: All green
01:20:47JdGordon|kugel: draw_viewport() just draws a viewport border?
01:23:31 Quit JdGordon (Ping timeout: 245 seconds)
01:24:48 Quit pamaury (Remote host closed the connection)
01:25:37kugelJdGordon|: yes
01:25:59JdGordon|can you rename it to "draw_viewport_border()" then please?
01:26:13kugelno, why?
01:26:22kugelit matches with drawrect()
01:26:22JdGordon|otherwise it could eazsily be confused with update_viewport()
01:26:37JdGordon|drawrect is obvious what it does
01:26:54JdGordon|draw_viewport_rect would even be fine
01:27:02 Quit stripwax__ (Read error: Connection reset by peer)
01:32:49kugelI'm not sure how that makes it less ambigious but well
01:33:57JdGordon|without looking at the code how is anyone supposed to know what draw_viewport actually does?
01:34:11JdGordon|at least draw_vp_rect (or whatever) gives a tiny bit of indication
01:34:27kugelif you say so
01:34:47JdGordon|I wouldnt have asked if I thought otherwise
01:35:08CIA-7New commit by 03kugel (r28240): Add _rect to {draw,fill}_viewport as suggested by Jonathan Gordon to reduce the chance to confuse it with update_viewport().
01:35:17gevaertsJdGordon|: seems to fix the Player progressbar issues
01:35:26JdGordon|kugel: thanks
01:35:29 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
01:36:24JdGordon|gevaerts: I'll take your word for it :) why move those functions out of skin_display.c though?
01:36:47JdGordon|I'd prefer them go back
01:37:19CIA-7r28240 build result: All green
01:37:43gevaertsJdGordon|: because I needed them in skin_tokens.c. The thing is that these are a bit in between. They provide format strings (well, the fullbar() one does), *and* they set up the display to be able to render them
01:38:17kugelnow we could write a button widget it probably less than 10 lines
01:38:31gevaertsThe rest of the code seems to do either getting a string (skin_tokens.c) or drawing something (skin_display.c)
01:38:38JdGordon|it wont beat the awesomeness of thr yes/no buttons :D
01:38:56JdGordon|gevaerts: ok, fair enough
01:39:28S_a_i_n_tpixelma: When a patch complains about ending unexpectedly, it usually just means whatever it was posted to mangled it by removing the newline at the end.
01:39:31gevaertsJdGordon|: I can move them back, but I don't really see how I can move calling them out of skin_tokens
01:40:30gevaertsWell, for %pb that might be doable, but for %pf it would be a mess I think
01:40:40JdGordon|ok, fine by me
01:40:47gevaertsok. Shall I commit?
01:43:56 Quit krabador (Ping timeout: 255 seconds)
01:46:23CIA-7New commit by 03gevaerts (r28241): Fix charcell %pb and %pf tags, FS #11592 ...
01:48:15CIA-7r28241 build result: All green
01:48:54 Join smtc96 [0] (
02:05:38 Quit Bagder (Ping timeout: 276 seconds)
02:08:16 Join Bagder [0] (~daniel@rockbox/developer/bagder)
02:08:18JdGordon|kugel: stokv tagcache building the db
02:16:14kugelJdGordon|: fix it :)
02:16:33kugelthe db scans the whole file system which is overkill in raaa
02:16:57JdGordon|we shuold use sql for tagcache on raa
02:18:30saratogalets use a non-transactional database
02:18:35saratogafor better scalability
02:18:36 Quit Gabe_G23 (Ping timeout: 255 seconds)
02:18:54linuxstbkugel: Are you sure about your implementation of those functions when the vp parameter is not null? Won't the lcd drivers be drawing in the current viewport, which may not be the full screen? Your original commit also seems to have some unrelated changes.
02:19:29 Join anewuser [0] (kvirc@unaffiliated/anewuser)
02:21:12 Join Gabe_G23 [0] (~gabe@bzflag/player/GabrielG)
02:21:35kugelit'll draw the passed viewport, relative to the current_vp. is that wrong? every drawing is relative to current_vp
02:21:59linuxstbYou mean a viewport within a viewport?
02:22:41kugeldo you think it should be absolute?
02:23:03linuxstbI thought viewports were always absolute. But maybe I've missed something in the last few months...
02:23:22kugeldraw_viewport_rect is equivalent to drawrect(0, 0, vp.width, vp.width)
02:23:39kugelit could also be that I missed something
02:24:03linuxstbAFAICS, you are only passing NULL to those functions at the moment?
02:25:28kugelJdGordon|: there's also update_viewport_rect() btw
02:25:57kugellinuxstb: there's no other drawing func which takes a viewport so I guess there's no convention yet
02:26:02JdGordon|which is what? update a rectangle inside the viewport?
02:27:01linuxstbkugel: Well, lcd_set_viewport() is an lcd driver function which takes a viewport parameter.
02:27:15kugelthat's no drawing function
02:28:32 Quit Gabe_G23 (Ping timeout: 245 seconds)
02:30:52linuxstbYes, I know. So what are you planning? To add more functions which take relative viewports? Your function also isn't taking account of the other values in the viewport, such as colours.
02:34:36 Join Gabe_G23 [0] (~gabe@bzflag/player/GabrielG)
02:34:46kugelI'm not planning anything
02:35:42linuxstbDo you actually need the case where vp != NULL?
02:36:05kugelnot yet
02:36:25linuxstbThen I would say remove it - don't code things you don't need.
02:36:56kugelthat'd be consistent with clear_viewport
02:37:16kugelalthough in the past I sometimes wished clear_viewport() would take a viewport
02:40:44linuxstbAs a relative or absolute viewport?
02:45:04kugelI don't know anymore, but probably absolute
02:46:35 Quit Gabe_G23 (Quit: Server lost the game. :()
02:47:36 Join Gabe_G23 [0] (~gabe@bzflag/player/GabrielG)
02:47:50kugelwithout the viewport parameter the _rect looks strange now
02:47:54linuxstbkugel: I also think more thought needs to go into the names - "display->fillrect(0, 0, button_width - 1, vp.height);" is self-explanatory when you read it, but "fillrect_viewport(NULL)" isn't IMO - it's not comparable to fillrect. Your original lcd_fill_viewport was better in that case...
02:48:46linuxstbmaybe lcd_drawborder_viewport for the other?
02:49:11linuxstbkugel: I was saying that due to the badly named functions, the code was easier to read before your commit.
02:49:40kugelin my initial commit it didn't have the _rect (which is why I just said it looks strange now)
02:49:59kugelit was fill_viewport(NULL) originally
02:51:04*JdGordon| didnt have a problem with fill_viewport, draw_viewport() on the otherhand was misleading
02:51:33linuxstbkugel: Yes, I'm agreeing with you - our messages crossed (i.e. I hadn't read your line before writing mine)
02:51:57linuxstbTo be clear, my suggestion would be lcd_fill_viewport() and lcd_drawborder_viewport().
02:52:57kugelmaybe lcd_fillrect_viewport() and lcd_drawrect_viewport()?
02:53:34*kugel would like fill_viewport and draw_viewport only but I must be the only one who doesn't see it's any more or less confusing with update_viewport()
02:54:06JdGordon|kugel: if /sdcard/rockbox/wps/ exists at all the inbuilt cabbie.wps wont work right?
02:54:19 Part smtc96
02:54:19JdGordon|bugger :p
02:54:34linuxstbkugel: I think we're all agreed on lcd_fill_viewport() unless I'm misunderstanding.
02:55:06kugelI think both should have a consistent name
02:55:32 Quit leavittx_ (Ping timeout: 264 seconds)
02:55:37kugelconsistent with each other and consistent with drawrect() and fillrect()
02:56:13*kugel hates that some of the older functions have no underscore between the words
02:57:13JdGordon|haha crap! dont enter the font browser in absolute mode!
02:57:14linuxstbkugel: I understand your argument. But I also think that lcd_draw_viewport() doesn't describe what it does very well - i.e. draw a border inside the viewport.
02:57:16JdGordon|you cant get out!
02:59:50kugeldrawrect isn't better in that regard
03:00:17kugelalthough I personally don't see what's wrong it, I'm not confused
03:00:42JdGordon|don't take this the wrong way, but of course you dont see the problem.. you implemeneted it
03:01:09JdGordon|and yes, the lcd funcitons are all rather messily named
03:03:48kugelI think draw perfectly describes what it does (especially when I translate it into my language), I connect it with drawing lines on a paper
03:04:29***Saving seen data "./dancer.seen"
03:04:45kugelbut well, I can see your point as well. drawborder_viewport then?
03:07:08JdGordon|that would be ok
03:07:27 Join dalinkwent [0] (
03:07:30JdGordon|I sort of wonder why this is so far down though
03:07:31linuxstbThinking about it some more, should they even have "viewport" in the name? The other LCD drawing functions don't.
03:09:33kugelthe text ones you mean?
03:10:08dalinkwentack: hey all, was looking for info on psgroove for the fuzev2 earlier and saw some comments saying you guys were looking for testers?
03:10:31kugelnot psgroove...
03:11:04linuxstbkugel: I mean all lcd functions, apart from the four that deal with setting/clearing/updating a viewport.
03:11:12dalinkwentwell i think the guys working on psgroove are maybe building off your recent work with usb in rockbox?
03:11:23 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
03:11:26kugelIMO clear_viewport is a drawing function
03:11:53kugelyou can implemented it with set_foreground and fill_viewport calls
03:12:27kugelset_drawmode, not set_foreground
03:12:57dalinkwenti have a fuzev2 with version r8228-101010 and if anyone needs anything tested im here to help out
03:13:06dalinkwentsorry if im interrupting
03:16:27S_a_i_n_t"testing" is basically using the most current build available, and reporting bugs, if any.
03:16:41S_a_i_n_tso, you're doing it just by using the device and keeping in touch.
03:17:03S_a_i_n_tdalinkwent: ^
03:18:04linuxstbkugel: Yes, but the only one. That could be "lcd-clear()". But anyway, I don't want to argue all night about names...
03:18:26kugelthanks :)
03:19:57CIA-7New commit by 03kugel (r28242): Rename {draw,fill}_viewport once more (to draw_border_viewport and fill_viewport) and remove the viewport parameter as suggested by Dave Chapman. ...
03:20:08dalinkwentwell i updated today with rockbox util, and i dont know if its exclusive to the newest build (havent updated in a while) but my fuze will now stay in rockbox when connected with usb, and it also shows that its charging in rockbox now
03:20:41S_a_i_n_tdalinkwent: did you select current or release build?
03:20:53JdGordon|kugel: draw_viewport_border would be better english :)
03:20:57S_a_i_n_tcurrent I'm thinking...and yerah, USB was implemented a while back now.
03:21:10dalinkwentyea i used current build
03:21:42dalinkwentwhich means im probably using an older version than whats being worked on?
03:21:57kugelnobody is working on psgroove for fuzev2
03:22:12S_a_i_n_tconsidering there's 10-to-50 commits per day, yes.
03:22:16S_a_i_n_tbut not by much.
03:22:20CIA-7r28242 build result: 148 errors, 26 warnings (kugel committed)
03:23:42CIA-7New commit by 03kugel (r28243): Forgot to update lcd-remote.h
03:24:12dalinkwentOriginally Posted by sk8r8sht View Post Over at rockbox this morning Ive noticed that they seem to have maybe found a workaround for the fuze v2 usb. This might be what alot of people are looking for. They currently need testers if anyone is interested, to get the ball rolling.
03:24:28dalinkwentthats basically the post that pointed me here
03:24:33dalinkwentthis is obviously over my head
03:24:38dalinkwentjust wanted to see if i could lend a hand
03:25:47CIA-7r28243 build result: All green
03:27:03kugelyou can test but it has nothing to do with psgroove
03:28:54 Quit kugel (Remote host closed the connection)
03:35:57 Quit dalinkwent (Quit: CGI:IRC)
03:36:13 Join webguest41 [0] (
03:37:41S_a_i_n_thmmmmm, slightly freaking out, but not really...does automatically build arm-eabi for arm toolchain now?
03:37:50S_a_i_n_tjust realised I typed a instead of e
03:38:08S_a_i_n_tand that e is no longer an option.
03:38:25JdGordon|I woudl say.. probably...
03:38:59S_a_i_n_tthat's what I thought, but it's *almost* I'd be pissed if I was wrong ;)
03:39:56JdGordon|hmm, we cant draw a plain rectangle in the skins currently can we?
03:40:09 Quit webguest41 (Client Quit)
03:40:14S_a_i_n_tdepends how you mean.
03:40:15JdGordon|a filled in rectangle I mean
03:40:28S_a_i_n_terrr...not that I know of.
03:40:54JdGordon|I might need to add access to fillrect and drawrect
03:41:16JdGordon|I want a light border behind my AA but apparently cant do that without loading an image'll need to load an image.
03:41:42S_a_i_n_tIt'll hardly cost a lot.
03:42:06JdGordon|that isnt the point
03:42:11JdGordon|I shouldnt need to :p
03:42:15JdGordon|who wrote this crap anyway?!
03:42:33S_a_i_n_tIf everyone else needs to, you do tyoo!
03:42:44JdGordon|on contraire
03:43:13 Quit BHSPitMonkey (Remote host closed the connection)
03:43:13JdGordon|%d is unused right?
03:43:19S_a_i_n_tpfffft! typical dev having his cake and eating it too, in front of the starving users ;)
03:43:57JdGordon|%dr(x,y,w,h,filled,[colour]) draw rect, optionally filled and coloured
03:44:23*S_a_i_n_t squints for something to complain about in there.
03:44:33JdGordon|you dont have to complain... :p
03:44:33S_a_i_n_tgimme some time, I'll find it eventually.
03:44:50S_a_i_n_tI don't *have* to ;)
03:45:00 Quit z35 (Quit: Leaving)
03:45:20JdGordon|ahem... %dr(x,y,w,h, filled?, start colour, end colour)
03:45:23JdGordon|graident fill ftw
03:45:47S_a_i_n_tis there a gradient thingo?
03:45:56S_a_i_n_tah...yeah, the line selector can do it.
03:46:11JdGordon|yeah, dunno how it works, but if it can do it there i should be able t use it
03:46:33S_a_i_n_t(famous last words)
03:48:16JdGordon|ahem... %dr(x,y,w,h, filled?, start colour, end colour, <value tag to use this as a bar>)
03:48:43JdGordon|maybe not :p
03:48:49 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
03:49:15S_a_i_n_tno, it'd be better (though I can't believe I'm saying this...really) to kludge that into the bar's line, not the other way around.
03:49:43S_a_i_n_tso the bar gets and additional: filled? how? param.
03:49:54JdGordon|yeah, probably
03:49:59JdGordon|although it already has enough params
03:50:05S_a_i_n_tthough, it's not like the bars need any more optional params ;)
03:50:34JdGordon|although I dont think it is unusable with all of them
03:56:39rasherJdGordon|: can't you just make a viewport with a background colour?
03:56:52rasherAlso, more features right before a release? Really?
03:57:07JdGordon|no, because the stupid backdrop image gets in the way
03:57:19JdGordon|and no, this is a bug
03:57:36rasherAren't viewports with background colours drawn on top of the backdrop?
03:57:41S_a_i_n_thow do you mean "backdrop image gets in the way"?
03:57:50S_a_i_n_twhat rasher said should indeed be possible.
03:57:56rasherAs an opaque rectangle, that is.
03:57:58S_a_i_n_tI feel stupid for not thinking about it.
03:57:59JdGordon|the background colour is actually ignored by the lcd code
03:58:07rasherWell *that* sounds like a bug to me
03:58:14rasherAdding a new tag, however, is certainly not
03:58:40S_a_i_n_tIt should "just work" adding a viewport behind the AA with a BG colour.
03:58:53S_a_i_n_tthe fact that it doesn't is the simpler fix here, and the real "bug"
03:59:03*JdGordon| isnt so sure
03:59:15*S_a_i_n_t is, if it doesn't's broken.
03:59:17JdGordon|I could make %Vb() do that, but that could break peoples themes
03:59:18S_a_i_n_tit used to.
03:59:36JdGordon|no, the background colour never affected anything if a backgorund image was loaded
03:59:47S_a_i_n_t*shouldn't* %Vb() do that?
03:59:55S_a_i_n_tisn't that what it's for?
03:59:55 Quit Rob2223 (Quit: Rob2223)
04:00:04JdGordon|well, sort of
04:00:16 Join Rob2222 [0] (
04:00:19S_a_i_n_tDoes it get ignored if there's a backdrop?
04:00:24JdGordon|it sets the viewport backdrop colour, the lcd driver is the actual problem here
04:00:25S_a_i_n_tif so, that's pretty silly.
04:01:00S_a_i_n_tAh, right...well, I see fixing %Vb() as the "real" issue here, personally.
04:01:12S_a_i_n_tIt will allow you to do what you want with no additions.
04:01:41JdGordon|gradient fill would still be nice (although easily added to Vb)
04:01:45S_a_i_n_tand no "pre-release-sneaking" that'll almost certainly cause a shitstorm.
04:01:55S_a_i_n_tAh, yes, it would be nice.
04:01:57JdGordon|being able to draw a empty rectangle has been something I've wanted for a while also
04:02:07S_a_i_n_tbut a working %Vb() would be nice too.
04:05:12rasherIf you set a background colour for a viewport, it should be filled with that colour
04:05:48JdGordon|it has never worked that way though if a bg image was loaded
04:05:58JdGordon|so yes, I agree 100%, it could break peoples skins
04:06:04JdGordon|although, its a dead simple fix
04:06:23S_a_i_n_tHuh, I never really played with it...I always load images as I very rarely, or apparently never have had a need for this.
04:06:27rasherWhy would people be setting a background colour anyway, if it's never actually worked?
04:06:35S_a_i_n_tbut the way rasher and I think it should work seems very sane.
04:06:40JdGordon|because it used to be required in the %V line
04:06:51rasherBut it allowed -, didn't it?
04:06:51S_a_i_n_trasher: It works, butnot if there's a backdrop.
04:06:53JdGordon|I'm agreeing
04:06:55S_a_i_n_tI do remember this now.
04:07:13*JdGordon| wants a dev env right now :p
04:07:35rasherI don't see why it should matter if there's a backdrop or not
04:07:55S_a_i_n_tIt shouldn't, but that's the way I remember it.
04:08:14JdGordon|ah, hmm... I dont tihnk this will actually work
04:08:28JdGordon|clearing a line, and the scroll engine wont work
04:09:03rasherI'm not arguing how it works now, just how I think it should work
04:09:04 Quit ajb (Read error: Connection reset by peer)
04:09:09*S_a_i_n_t looked at the scroll engine, once...
04:10:10 Quit amiconn (Disconnected by services)
04:10:11 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:10:29 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:10:49 Quit pixelma (Disconnected by services)
04:10:51 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:10:53 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:12:12JdGordon|yeah, the lcd driver needs work for this to be fixed properly
04:13:22 Quit TheSeven (Ping timeout: 265 seconds)
04:13:24 Join DerPapst1 [0] (
04:13:37S_a_i_n_tWhat's the actual issue with the LCD driver?
04:14:08JdGordon|it ignores the viewport bg_pattern if an image is loaded
04:14:22linuxstbMaybe have something like a "show_backdrop" flag in the viewport struct?
04:14:30 Quit DerPapst1 (Client Quit)
04:14:33linuxstbOr rather, if a backdrop is loaded, all viewports will use it.
04:14:34JdGordon|so everything that draws transparenrlty, or clears will show the wrong thing
04:14:58JdGordon|linuxstb: the latter is what is happening now
04:14:59 Quit DerPapst (Ping timeout: 265 seconds)
04:15:09S_a_i_n_tHmmm, if this *is* fixed, would it not allow loading a backdrop for each individual viewport?
04:15:14S_a_i_n_tthat could be kinda cool.
04:15:18JdGordon|it wouldnt
04:15:33S_a_i_n_tbah, oh well...dreams are free.
04:15:34JdGordon|seperate issue, and annoyingly a moslty nodo
04:16:03linuxstbJdGordon|: Yes, I know - I was just describing the current situation a different way.
04:16:10S_a_i_n_tit's not the loading that's the issue though is it, it's where it's stored. Correct?
04:16:29S_a_i_n_t(re: multiple BGs)
04:16:52S_a_i_n_tDynamic skin buffer FTW
04:17:09 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:18:05JdGordon|multiple bg's are difficult because the way the correct pixel is got to is very simple math, having seperate images means slightly more complex math being called very very often
04:18:44JdGordon|it could be done, but I tihnk there is actually opposition to it
04:19:02JdGordon|adding a flag to make the vp ignore the bg image should be doable though
04:19:12S_a_i_n_tAs there is for most of the cool things that *could* (in theory) be done with the skin engine.
04:27:10 Join Chronon [0] (
04:27:12rasherJust because it can be done doesn't make it a good idea
04:27:55JdGordon|I disagree (both with that statement in general, and in using it with this idea)
04:28:25linuxstbJdGordon|: You're saying that there are no bad ideas?
04:29:16JdGordon|discussion on that doesnt belong here :)
04:30:05LloreanCorrection then: "You're saying there are no bad ideas for Rockbox changes?"
04:34:36 Quit S_a_i_n_t (Quit: I'm only going to Heaven if it tastes like caramel...)
04:34:50 Join Judas_PhD [0] (
04:34:51rasherJdGordon|: So you're saying there's no reason we shouldn't include anything someone can think of?
04:35:03 Join S_a_i_n_t [0] (S_a_i_n_t@
04:37:13 Quit MethoS- (Remote host closed the connection)
04:39:47JdGordon|I didnt say that at all
04:41:40JdGordon|and anyone that has ever read irc or the ml knows I do entertain new ideas more freely than (apparently) lots of other people (not saying that is good or bad)
04:42:07*linuxstb doesn't know what we're arguing about
04:42:07*S_a_i_n_t personally has seen many of his "wild ideas" come to fruition.
04:43:21S_a_i_n_tIn fact, "next AA" is the only thing on my list that JdGordon| hasn't checked off...and every single one of those was "na, I don't think so" at some point.
04:46:00rasherI'm just saying that there may be very valid reasons to reject an idea.
04:46:19rasherDisagreeing with that seems to mean "every idea should be implemented", basically.
04:46:26JdGordon|anyone want to test viewport backgrounds on a grey target with a wps that does set %X?
04:46:43*JdGordon| is curious as to what happens there
04:46:59S_a_i_n_tit doesn't mean every odea should be implemented, but, it should be considered without a closed mind.
04:47:06S_a_i_n_tI had to really push for some of my ideas.
04:47:14rasherI didn't say that.
04:47:51S_a_i_n_tadding the "basically" made it different?
04:49:19rasherI have no idea what you and JdGordon| are saying.
04:50:02 Join dys` [0] (
04:50:58JdGordon|all I'm saying is no idea should be rejected outright with little more than a sentence explaining it
04:51:03S_a_i_n_tWell, I'm saying personally that everything should be considered. Instead of immediately dismissing something as "a stupid idea" if it initially seems so.
04:51:31 Join Barahir [0] (
04:51:31S_a_i_n_tI'm sure there's been plenty of past "stupid ideas" that are now parts of RB
04:52:35rasherWho the hell said anything about outright rejecting ideas
04:53:17 Quit dys (Ping timeout: 276 seconds)
04:53:22JdGordon|calm down... thats how I interpretted "<rasher> Just because it can be done doesn't make it a good idea"
04:54:07rasherI'm perfectly calm, just incredulous as to why you're arguing a point no one made
04:54:48 Quit Barahir_ (Ping timeout: 250 seconds)
04:56:53JdGordon|I think we are both misinterpretting eachother... so lets drop it :)
05:03:37*JdGordon| adds a bug for the viewport ignoreing bgcolour
05:04:30***Saving seen data "./dancer.seen"
05:18:41 Quit anewuser ()
05:19:06 Quit Chronon (Ping timeout: 265 seconds)
05:34:53 Quit ps-auxw (Ping timeout: 276 seconds)
05:40:29 Quit edboyer93 ()
05:45:34 Join ps-auxw [0] (
05:53:58 Join Nausicaa [0] (
06:08:35S_a_i_n_tpixelma: amiconn: any other CygWin
06:09:57JdGordon|not at 5am I would bet :p
06:10:04S_a_i_n_tWho feels like being a dear and either packagine the arm-eabi toolchain for me, or opening an FTP so I can steal it from them?
06:10:23S_a_i_n_thence the "any other CygWin users" bit ;)
06:10:40S_a_i_n_tand, I *have* seen amiconn here at some odd times.
06:11:14JdGordon|remind me when I get home and I might be able to help...
06:11:28JdGordon|although you actual upload wont be able to happen till tomorow when I'm at work again
06:11:50S_a_i_n_tyou have a CygWin env?
06:12:11JdGordon|no, but I can set one up easily enough...
06:12:21S_a_i_n_tyou'd think.
06:12:31S_a_i_n_tI seem to be failing at it.
06:12:35JdGordon|well, comp is fast enough to not worry about building toolchains :)
06:12:51S_a_i_n_tthe package manager has changed, and it was always VERY vague what was needed to make eabi
06:13:15S_a_i_n_tI bluffed it last time, there's many a package the wiki misses.
06:13:29S_a_i_n_tI should have fucking updated it while I knew them all.
06:14:18S_a_i_n_tI have my doubts that it failed solely due to a failed rm action
06:14:35S_a_i_n_tlibiconv was playing some asshole games with me the last I rememebr
06:21:59JdGordon|anyone got a radio target? is .rockbox/wps/rockbox_failsafe.fms created?
06:36:10 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
07:03:03 Quit Judas_PhD (Quit: This is a quitting message)
07:04:32***Saving seen data "./dancer.seen" going to mess up booting the latest build with the older bootloader?
07:06:12S_a_i_n_tI trust a new bootloader will be released?
07:07:31S_a_i_n_tI'm thinking it may mess things up because it touches more than the bootloader, the clickwheel driver is modified also and I'm not sure if it is backward compatible or not.
07:14:50S_a_i_n_tis anyone aware what the issues;revision=28164 introduced were? (the reason it was reverted)
07:15:22S_a_i_n_tI' going over past commits, and this one isn't very well commented.
07:16:29S_a_i_n_tI do know that the annoying hum that the LCD makes on Nano2G is bloody annoying, and it does diminish if the voltage is decreased...I never saw any issues with the screen in my own trials.
07:17:34S_a_i_n_tWith the current voltage playing rockboy for 1-to-2 hours + makes the LCD make a horrible "I'm going to die soon" crackling noise, seems to happen when the CPU is boosted for long periods.
07:20:09 Join feisar- [0] (
07:34:38*S_a_i_n_t needs theme testers for iPod Video, any takers?
07:35:25 Join G4Oblivion [0] (
07:36:36S_a_i_n_tG4Oblivion: Any more luck debugging that annoying panic on Nano2G?
07:36:47G4OblivionS_a_i_n_t: The reason the commit was reverted was because a user in forums said that his nanos screen would flicker and go blank. I'm unsure if they confirmed that it was a ILI9320 LCD type problem.
07:37:15G4OblivionI deleted my config and am using the Rockbox default right now. I have not had any problems since.
07:37:36S_a_i_n_tG4Oblivion; Ah, I see...I saw that the commit was re-commited again later, so I guess it wasn't confirmed.
07:38:01S_a_i_n_thave you tried skipping *really* fast in the WPS?
07:38:10S_a_i_n_tI find skipping backward seems to do it for me.
07:38:29S_a_i_n_tIt does it skipping forward also, but backward seems to do it a lot quicker.
07:39:10S_a_i_n_tIt seems to be an issue with buffering, I haven't tried reverting my config file however, I'll do that now and see if I still get the problem.
07:40:38S_a_i_n_tG4Oblivion: Are you able to test a theme on your Nano2G for me?
07:40:58G4OblivionSure. Just tell me what to look for.
07:43:15S_a_i_n_tI'm basically just lookinf for any visual flaws (if any), but it would also be nice if you could test if fallbacks for cases without metadata are handled correctly.
07:43:26S_a_i_n_tAnd, anything you feel could be improved.
07:43:36S_a_i_n_tfile is here:
07:43:46G4OblivionWill do.
07:44:37S_a_i_n_tAnd, after testing it'd be good if you could PM me a real name so I can add you to the credits.
07:45:45G4OblivionOk. Just to make sure metadata is track info right? anything else metadata related?
07:47:06S_a_i_n_twell, it *should* be displaying folder level 3 for album, and folder level 2 for artist. It assumes a sane folder layout. ie: Music - Artist - Album
07:47:17S_a_i_n_tJust wanting to make sure that's working correctly.
07:48:03S_a_i_n_tif it can't display those, it will display bitrate and samplerate IIRC
07:48:43S_a_i_n_tbut, if you have the correct metadata present, it should just display that.
07:49:48S_a_i_n_tI also want to check if the "next track" queues are displaying correctly as I was having trouble with that on my units here. You should get a "Next Track" prompt display Artist and Album 15 seconds from the end of the current track.
07:51:45G4OblivionStill testing but, isn't the scrollbar color a bit dark if its supposed to look like OF?
07:53:13G4OblivionHmm...Is song named supposed to be fixed to the right side instead of the middle like artist and album
07:53:40S_a_i_n_tNo, I guess it isn't.
07:53:56S_a_i_n_tby scrollbar, do you mean the list scrollbar?
07:54:19S_a_i_n_tI can't really do anything about that until there's bitmap scrollbars available.
07:54:31S_a_i_n_tthe theme.cfg should disable scrollbar.
07:55:08 Quit Nausicaa (Ping timeout: 245 seconds)
07:55:54G4OblivionOk. On long track names it displays the name correctly. On short track names that don't need to scroll is has it fixed to the right side of the screen instead of the center.
07:56:02S_a_i_n_tHmmm, looking at the code, that alignment issue shouldn;t be happening.
07:56:10S_a_i_n_tthat's odd. thanks for the headsup.
07:56:40S_a_i_n_tIs this with or without AA?
07:56:47S_a_i_n_t(album art)
07:57:10G4OblivionI don't have any album art tracks to test =/
07:57:15S_a_i_n_tvery weird.
07:57:27S_a_i_n_tand, that's fine..AA isn;t necessary.
07:58:19G4OblivionNext track appeared at -0:08 instead of -0:15
07:58:20S_a_i_n_tout of curiousity, is the language on the player left to right (as in english) or right to left?
07:58:33G4Oblivionleft to right
07:58:50S_a_i_n_tOk, that's least it's popping up. I assume you have crossfade set?
07:59:06G4OblivionOnly if its on by default.
07:59:20S_a_i_n_tcrossfade seems to mess with the %pE/%pS settings
07:59:30JdGordon|it does?
07:59:51JdGordon|fun :p
08:00:02S_a_i_n_tG4Oblivion: Oh, should be appearing at -10, not -15. My mistake
08:00:14S_a_i_n_tbut either way -8 is wrong ;)
08:00:34S_a_i_n_tthe alignment issue is a fun one...the code doesn't give me any clues why that is happening.
08:00:36G4OblivionJust tested again. It appeared at -0:10 correctly.
08:03:20S_a_i_n_tOk, I just fixed the alignment issue...stupid typo/forgot to remove old code.
08:03:38G4OblivionEverything seems to be working with normal use.
08:03:52S_a_i_n_tG4Oblivion: Could you check if the titlebar is displaying charging status correctly please?
08:04:16S_a_i_n_tI have had problems with that here, and my dev environment is down at the moment so I can
08:04:28 Quit BHSPitMonkey (Ping timeout: 264 seconds)
08:04:28S_a_i_n_t't test on a new enough build.
08:04:51S_a_i_n_tThe build on my iPods is about a month old.
08:05:05G4OblivionI've never tried charging through rockbox. Do I need to turn on charging and plug it in?
08:05:10JdGordon|you can download from the interwebs!
08:05:33S_a_i_n_tnot with all my patches I can't ;)
08:05:51S_a_i_n_tRB seems terrible to me without the setup I like. (no offence)
08:06:19S_a_i_n_tG4Oblivion: Charging *should* be anabled by default.
08:06:30JdGordon|i dont tihnk it is actually...
08:06:39S_a_i_n_tYou might need to set it to "force" depending on the port you're charging through.
08:06:42JdGordon|uh, it will connect but only at 100mA i think
08:07:03S_a_i_n_tAh, "force" tells it its safe to grab 500 at all times.
08:07:05G4OblivionIs there a way to exit usb mode without hard reset?
08:07:15S_a_i_n_tBut you need to know it actually *is* safe to do so.
08:07:27S_a_i_n_tunplug USB
08:08:00S_a_i_n_tIf you unplug USB and it stays on the USB screen something is horribly wrong
08:08:03G4OblivionI did still on USB screen. My computed didn't recognize it when I plugged it in either.
08:08:37S_a_i_n_tthat's usual for the Nano2g, did it just not enumerate? or did it fail to mount completely?
08:08:46S_a_i_n_tUSB HID is fucked for the Nano2G
08:09:21S_a_i_n_tI mean, it certainly is not normal for it to stay on the USB screen if no USB is present.
08:09:22G4OblivionAfter reboot its working. Is it because I had a song going when I pluged it in?
08:09:26S_a_i_n_tthat's definitely a bug.
08:09:35S_a_i_n_tIt shouldn;t be.
08:09:52G4OblivionUSB was still plugged into iPod but not computer. I may be at fault.
08:09:53S_a_i_n_tIt should have just stopped playback.
08:10:10S_a_i_n_tYou can also hold Menu to charge and continue use/playback
08:10:44S_a_i_n_tthat still seems wrong, plugging USB in, but not having it plugged to the PC should do nothing.
08:10:55 Join Na|L [0] (
08:11:05 Quit JdGordon| (Quit: leaving)
08:11:05G4OblivionI see "Charging" "Current Time" and a flashing charging battery symbol.
08:11:09S_a_i_n_tit should only know a cable is there when it detects power.
08:11:23S_a_i_n_tyep, cool...that works then.
08:11:37S_a_i_n_tit should switch to "Charged" when the battery is full.
08:12:00G4OblivionMines almost dead =P
08:12:23S_a_i_n_tAsides from obvious differences, you think I've got the look pretty close?
08:12:44S_a_i_n_tThe aim is (eventually) to not be able to tell whether it is the OF or RB
08:12:53Na|Lhello, anybody iriver h3xx positive? simple question about restart while charging
08:12:57G4OblivionI haven't used OF in months but it seems to be pretty close.
08:13:17S_a_i_n_tonly thing left is the scrollbar, and that's not possible at present...when bitmap scrollbars can be done it should be pretty much perfect.
08:13:41S_a_i_n_tNa|L: It's best to just ask your question.
08:14:03 Join n1s [0] (
08:14:10 Quit n1s (Changing host)
08:14:10 Join n1s [0] (~n1s@rockbox/developer/n1s)
08:14:34G4OblivionText is a bit choppy. A smoother text would be nice. I'll boot into OF ad look at the differences. I can't test the OF WPS though, I have my music in my own folder not iTunes.
08:15:55S_a_i_n_tyeah, I can't really do anything about the text unfortunately. that's as close as I can get it until/or if there is ever, aliased fonts in core.
08:16:18Na|Li put usb cable in wrong hole and result is: some smoke and player stay in "charging state" even if not, so i cant restart/poweroff player, my question is, bloking poweroff is made by SW or it's HW block?
08:16:25S_a_i_n_tI don;t quite have the motivation to do a custom build with aliased fonts enabled just for this, maybe one day.
08:17:58S_a_i_n_tNa|L: Sounds like you blew something up I understand it the power is detected by hardware and controlled by software
08:18:30S_a_i_n_tbasically, it is recieving incorrect info now, and thinks its always charging.
08:18:40 Join JdGord [0] (~jd@
08:19:22S_a_i_n_tYou'll need to replace whatever it is you blew up, and with a bit of luck it should return to normal function.
08:19:56Na|Li know i blew something probably lx2201 ...
08:20:24S_a_i_n_tHow did you manage to put the charging cable in the wrong socket?
08:21:05S_a_i_n_tI thought they were different sizes so you specifically couldn;t do that...been a long time since I've seen an h3xx though.
08:21:17Na|Lthat was standard cable, i put it in media socket lot of time, no problem, but one day .... :)
08:21:41S_a_i_n_tmedia socket?
08:22:11S_a_i_n_tI'm asking, what is a "media socket"?
08:22:28Na|Li try make some modification to source to set state or charging permanently to "discharge" but withou luck :/
08:22:35G4OblivionAll the repeat and shuffle icons seem to appear correctly. Seems the OF has small spaces before the text on the home screen (i.e Music, Photos, Extras.)
08:23:01Na|Laaaaa, when player act as usb host, for downloading photos from digital camera etc
08:23:07G4OblivionIs it possible to add blank icons or something?
08:23:32S_a_i_n_tG4Oblivion: Good idea.
08:23:45S_a_i_n_tI hadn't thought of that actually, nice one ;)
08:24:28S_a_i_n_tOh, actually...I tell a lie.
08:24:50S_a_i_n_tI have tried that and it looked weird because the line selector didn't go to the left edge.
08:25:05S_a_i_n_tThat's another thing I can only fix with a custom build.
08:25:23S_a_i_n_tI'd need to add a space to the beginning of each menu entry.
08:25:57S_a_i_n_tie. so it was " Settings" as opposed to "Settings" in the .lang file
08:27:08S_a_i_n_tNa|L: I was unaware that you were able to charge the device through the devices host USB socket.
08:27:22G4OblivionHmm. I can't think of anything else.
08:27:39S_a_i_n_tNa|L: I guess it may not have had anything to do with what socket you plugged in, perhaps it was just the hardwares time to fail.
08:28:41S_a_i_n_tG4Oblivion: Thanks. Can you PM me a real name (preferably yours ;)) so I can add you to the credits please? I appreciate you doing this for me.
08:29:16G4OblivionSure. I just right click your name and select "Open Private Chat" right? I never us IRC clients.
08:29:35Na|Lmaybe, anyway thx.
08:33:06JdGordYou can put a blank in with the list icons
08:34:13S_a_i_n_tNa|L: Can you reboot the device?
08:34:47S_a_i_n_tYou should be able to hold "Menu(?)" and even if the player thinks it is charging you should be able to use it.
08:35:00JdGordSort of.. maybe
08:35:00JdGordEspecially if you sent using the largest icon size
08:35:00JdGordI think using 24*24 icons with appropriate transparency and centering will work with smaller line heights
08:35:47Na|LS_a_i_n_t: just by reset button(little hole), no by poweroff butten
08:36:03S_a_i_n_tJdGord: The problem is I need the line selector to go from one edge to the other.
08:36:15S_a_i_n_twith icons, or transparency it won;t do that.
08:36:36JdGordAh. Yes
08:36:49S_a_i_n_tNa|L: It is possible to use the player while charging by holding a button during charger connect.
08:37:11 Join antil33t [0] (
08:37:12S_a_i_n_tIn theory, you should be able to hold this button while it boots to get it to boot into RB properly.
08:37:34 Nick dys` is now known as dys (
08:37:41 Quit G4Oblivion (Quit: ChatZilla 0.9.86 [Pale Moon 3.6.9/20100913113105])
08:38:02n1sS_a_i_n_t, Na|L : afaik, the h300 bootloader always boots the OF if it thinks a charger is connected
08:38:56S_a_i_n_tquite possibly, but it is possible to hold a button (different for each target) that allows you to use the player while it charges.
08:39:04Na|Lplayer work ok, no problem, problem is just i cant poweroff it, or it never go to standby, so batteryfiletime is ....
08:39:34S_a_i_n_tdoes it charge still?
08:39:39n1sNa|L: well, that definitely sounds like hw failure
08:40:11S_a_i_n_tn1s: Smoke coming from the player usually means HW fail, yes ;)
08:40:27Na|Li just need knoe if it's posible do SW reboot/poweroff, even if HW thin it's on charger
08:40:47S_a_i_n_tAFAIK, no.
08:41:07S_a_i_n_tit will always say "charging" and prevent poweroff.
08:41:16n1swith luck, you might be able to hack the pcf driver to make it ignore whatever poweron signal it thinks it gets provided that it isn't one you actually need
08:42:00S_a_i_n_tbut then it will probably kill charging.
08:42:03 Quit JdGord (Quit: Bye)
08:42:06S_a_i_n_tSo you'd need a way to toggle it.
08:42:18Na|Lat least i need restart option in menu or something like that
08:42:20n1si remember preglow once managed to mess up the pcf config on some player and that wouldn't boot at all then :) (rescued by disconnecting the battery IIRC)
08:43:02n1sNa|L: if the pcf thinks a charger is connected it will start the player as soon as you poweroff
08:43:14S_a_i_n_tNa|L: Even if you have a reboot/poweroff option in the menu rockbox won;t let you reboot/poweroff if it thinks the player is charging.
08:43:35S_a_i_n_tthere is a splash that warns about charging, and it won;t let you reboot or poweroff.
08:43:44Na|Lthat sux :/ origin firmware do that, no poweroff but restart ist posible :(
08:43:44n1syou can use rolo to fake reboot
08:44:00S_a_i_n_tHmmm, true.
08:44:02n1swhy do you need to reboot anyway
08:44:07S_a_i_n_tRockbox Loader
08:44:22Na|Ldatabase update always want to reboot to "take efect"
08:44:56S_a_i_n_twell, add a second Rockbox image to the disk, and "run" it from the filebrowser.
08:45:02S_a_i_n_tthat will "fake" a reboot.
08:45:20n1syou don't need a second image, you can reboot the same one if you like
08:45:31S_a_i_n_tAh, good point.
08:47:26 Join leavittx_ [0] (~lev@
08:48:33 Quit n1s (Quit: Lmnar)
08:49:14S_a_i_n_tNo problem, good luck in replacing the hardware you blew.
08:49:34Na|Lit works, i named it reboot.iriver :D
08:49:35S_a_i_n_tit shouldn;t be too difficult if you have a heat gun or a very fine nibbed soldering iron.
08:50:38Na|Ldo it is no problem, i hack my openmoko, but buy that chip is problem, i wrote some emails, but no answare yet :/
08:50:55 Join ender` [0] (
08:55:14 Quit bluebrother (Disconnected by services)
08:55:15 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
08:57:35 Part Na|L
09:04:33***Saving seen data "./dancer.seen"
09:10:24 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
09:10:29 Quit leavittx_ (Ping timeout: 264 seconds)
09:11:24JdGordonamiconn: you know the lcd drivers right?
09:18:22 Join hebz0rl [0] (
09:18:47 Join p3tur [0] (d408b802@rockbox/developer/petur)
09:28:30 Join Zagor [0] (
09:28:30 Quit Zagor (Changing host)
09:28:30 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
09:34:57 Part pSiKO
09:38:02 Join Rob2223 [0] (
09:41:45 Quit Rob2222 (Ping timeout: 276 seconds)
09:43:52S_a_i_n_tCan an admin remove the link in please?
09:44:05 Quit JdGordon (Ping timeout: 264 seconds)
09:44:08 Join pSiKO [0] (~tbox@
09:44:15S_a_i_n_tthe theme is nowhere near as cleverly done as mine, and sues images pulled straight from the OF
09:45:01 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
09:45:33*S_a_i_n_t should put his theme for iPod Video up on the themesite to satisfy the needs of people that want an OF-ish theme
09:47:22S_a_i_n_twell..."A complete rip-off/duplicate/mockery of the OF theme" ;)
09:48:33JdGordondoes the sim use lcd-16bit.c or lcd-16bit-vert.c?
09:48:38S_a_i_n_tMine's way better, *and* I did it without using Apple graphics, though the almost might as well be as I certainly can't tell them apart anymore, there's a reason why I painstakingly recreated them as opposed to pulling them from the OF
09:52:35 Join swilde [0] (
09:54:49 Nick p3tur is now known as petur (d408b802@rockbox/developer/petur)
09:59:09JdGordon any guesses which lcd functions I've missed? I'm trying to get no background image
09:59:28 Join sasquatch [0] (
10:03:04 Quit utanapischti (Ping timeout: 245 seconds)
10:03:19JdGordonwoops, some ||'s inmstead of &&'s :p
10:04:22JdGordonthe transparent images still no worky, buut the text is
10:06:17JdGordonok, that was easier than expected :)
10:06:34 Quit TheSeven (Ping timeout: 240 seconds)
10:06:57S_a_i_n_tabckground colours work in viewports when there's a backdrop loaded now?
10:07:59JdGordonnot yet, one lcd driver is fixed so it works
10:11:11JdGordonwoo! butt ugly progressbar
10:11:29S_a_i_n_tcabbie? ;)
10:11:31JdGordondinner then fix the other drivers
10:11:47JdGordoncabbie with red pb in the unfilled area
10:13:33 Join earcar [0] (
10:17:11S_a_i_n_tamiconn: I posted in #rockbox-community, but I suppose it's more relevant here as there's a chance someone else maysee it also and be able to help. Is there a possibility you might be able to put the CygWin arm-elf-eabi toolchain up somewhere I might be able to grab it please? I would really appreciate it if you could do so, being one of the only other people I know that use CygWin dev environment.
10:20:15S_a_i_n_tI'm about 99% sure it isn't necessary to package it as a CygWin package, just pulling the directory structure out and compressing it (not sure it's really that big to begin with actually) would work just as well.
10:34:58 Join DerPapst [0] (~Alexander@
10:42:24S_a_i_n_tassuming the coding is correctly formatted, and sane, could someone please look at committing FS #11632 - (rockboy) Some fixes and improvements?
10:43:09S_a_i_n_tthe playback menu is not so usefull, but a few targets can potentially pull it off. the other improvements are all good though, and quite sane IMO
10:46:56pixelmaIIRC it shouldn't use the usual playback menu, it's about the game sound volume (I believe it was discussed here)
10:47:44*S_a_i_n_t looks again.
10:48:39S_a_i_n_t- I added a playback-control at the menu. Primary to change the ingame volume at rockboy. In some cases it should also be possible to play music while running rockboy, but on my player (sansa fuze v2), rockboy crashes then because there is not enought memory. Maybe somebody finds a way to catch this and makes something, that only shows the playback-menu on players, where both - gnuboy and music playing - is possible on the same time and where not only a
10:48:45S_a_i_n_tsorry for the huge post.
10:49:29S_a_i_n_tWhat he describes will only take some simple ifdeffing also, its safe to assume none of the lowmem targets will have a hope in hell of achieving playback+rockboy
10:50:08S_a_i_n_tthe beast and NAno2G should be able to do playback+rockboy
10:51:10pixelmawell, wouldn't your simple ifdeffing mean there won't be a playback menu at all on those players and then you again can't change the ingame volume? Isn't this what the author wanted?
10:51:49S_a_i_n_tNo, I mean you could remove all but the volume from the playback menu.
10:51:58S_a_i_n_tI probably should have been more descriptive
10:52:49pixelmathat would be very odd and I wouldn't expect to find ingame sound control in a playback menu
10:53:19S_a_i_n_tthis is true, it could be ranamed in those cases though.
10:54:00S_a_i_n_tthat would mean complicating the patch a LOT more than it is presently though.
10:54:44S_a_i_n_tthe [;ayback menu should be dropped from it completely IMO, the rest of the options the patch adds are quite useful though.
10:55:57S_a_i_n_tOnly a very simple change to remove the playback menu, which would more than likely increase the chance of it being cmmitted
11:04:35***Saving seen data "./dancer.seen"
11:04:57 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
11:14:12JdGordonwhich targets use which lcd drivers?
11:14:19*JdGordon is going to need help testing :(
11:15:39 Join Thanatos [0] (
11:18:23 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
11:30:24 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:30:30 Join kugel [0] (
11:30:32 Quit kugel (Changing host)
11:30:32 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:31:26kugelJdGordon: It's not a bug that the bg color is ignored when a backdrop is set
11:31:29 Join swilde [0] (
11:31:46 Quit hebz0rl (Quit: Leaving)
11:31:50JdGordonsee the log from a few hours ago
11:32:01JdGordons/few/8 or so/
11:33:54 Join openwound [0] (~g@
11:34:15openwoundcan i ask some rockbox?
11:34:56openwoundi have a mp3; model - sansa view
11:35:34 Quit kugel (Ping timeout: 245 seconds)
11:35:58openwoundi really want to adapt to my mp3; sansa view
11:36:09 Quit GodEater (Read error: Operation timed out)
11:36:09 Join kugel [0] (
11:36:11 Quit kugel (Changing host)
11:36:11 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:37:33kugelJdGordon: what are you referring to?
11:38:11JdGordonkugel: we discussed the bug today here.. it is a bug
11:38:12Stummiopenwound, afair there is not any done for sansa view yet
11:38:13 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:38:18amiconngevaerts: I'm only reading this now; the issues with plugins running greylib on COP sould like caching issues. The old plugin loading code was carefully designed to handle this
11:38:36amiconnGreylib is special because it's running an ISR on COP, not a thread
11:38:59kugelJdGordon: yes, I saw that. and I tell you it's not a bug
11:39:03pixelmaopenwound: what do you mean with "adapt"? Do you want to put work into porting, coding?
11:39:24kugelJdGordon: yes, I saw that. and I tell you it's not a bug
11:39:36JdGordonkugel: ok, fine, thats your opinion
11:40:04JdGordonwhich you are of course entitled to
11:40:53openwoundsansa view is abandoned?
11:40:56kugelthere can only be one background. bg color or backdrop. there is no such thing like two backgrounds
11:40:59JdGordondo any of the 2bit lcd targets have a cabbie with a background?
11:41:48kugeland what you plan seems to rather abuse it as a second foreground
11:42:05openwoundsansa view's developments are ceased.. right?
11:42:11JdGordonthis is NOT about more than one background, this is about forcing a viepwort to not use it
11:42:19JdGordonopenwound: noone is currently working on it
11:47:29amiconnJdGordon: Many of them, yes
11:47:35pixelmaJdGordon: sure (about greyscale cabbiev2 with background) - the now playing bar and the logo in the bottom left corner are part of the backdrop usually, the slightly bigger displays also have a menu backdrop
11:48:08JdGordonyeah, ok.. the 2 bit driver seems to be a bit trickier
11:49:05kugelJdGordon: there was once a feature request about what you now declare as bug
11:49:08JdGordonwhy is there a clearimgpixel() and clearpixel()
11:49:14kugelJdGordon: there was once a feature request about what you now declare as bug
11:50:15JdGordonoh, same as 16bit
11:50:22 Join GodEater [0] (
11:50:22 Quit GodEater (Changing host)
11:50:22 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
11:54:02amiconnA backdrop is always global and overrides bgpattern
11:54:08 Quit Dreamxtreme (Quit: Don't follow me)
11:54:21amiconnThis is done by using a different set of pixel and block functions
11:55:16amiconnThis is something I want to change, but it needs to be done efficiently
11:55:24S_a_i_n_tif someone doesn;t want the BG colour to override/overdraw the backdrop...they can easily eacape it with a -
11:55:55JdGordonno... the only way to bypass the image is in the skin using %Vb() which is what it should do
11:56:24S_a_i_n_tI mean, with the intended change
11:56:36JdGordonso do I
11:56:56S_a_i_n_tIf you don't want the BG colour to draw in a viewport...if a backdrop is loaded, using a "-" should be fine.
11:56:59JdGordonnowhere else in the code (well, currently) would override it
11:57:37S_a_i_n_tIf the BG colour will always draw over the backdrop..then this is going to suck ass.
11:57:50S_a_i_n_tperhaps I'm getting confused though
11:58:06kugelsee, it's not a bug. it's just not implemented
11:58:07JdGordonno, that is backwards... %Vb() is optional, only used if you want a solid colour instead of the image
11:58:27JdGordonamiconn: what do you want to change? or to how?
11:58:29openwoundi was wondering why Rockbox is not on the View
11:59:14S_a_i_n_topenwound: because you haven't done the work yet.
11:59:14JdGordonS_a_i_n_t: s/perhaps//
11:59:14JdGordonkugel: semantics
11:59:30 Part LinusN
11:59:59openwoundyet? 'yet' means that view is in progress?
12:00:12JdGordonnoone is working on it
12:00:21S_a_i_n_tno, "yet" means..."you haven't done the work yet"
12:00:56S_a_i_n_tBut, if you're interested in it, and have the seem like the perfect person to start.
12:02:08 Quit antil33t ()
12:02:51AlexPopenwound: See for the port process in general and for what is known about the View. As others have said, nobody is working on it.
12:03:25 Quit timccc (Quit: Leaving.)
12:03:37S_a_i_n_tThe work is far from trivial however.
12:07:12JdGordonhow does look? for some reason the list icons arnt drawing properly on the 2bit lcd
12:09:06JdGordon is that because the image source isnt transarent?
12:09:22 Join timccc [0] (~tim@
12:10:46 Join krabador [0] (
12:11:22pixelmaprobably - greyscale has no transparency. The only thing that comes close is using monochrome bitmaps then you have background/foreground colour
12:11:36pixelmahow does the default icon set look like?
12:12:21 Join LinusN [0] (~linus@rockbox/developer/LinusN)
12:12:22pixelmaguess it'll look fine though, seeing that text works
12:13:05JdGordonthe source image looks b+w to me, but it could be 2bit?
12:13:11 Quit timccc (Quit: Leaving.)
12:13:17JdGordonyeah text and the statusbar works so its probably ok..
12:13:25JdGordon2 more drivers to go.. and then... TESTING!
12:13:48pixelmathe tango icons? Not b/w at all
12:14:45JdGordonok, so that would explain it then
12:15:24pixelmalooks like the source image is named wrong
12:16:42 Quit openwound (Quit: Http://www.ZeroIRC.NET Zero IRC Ver 2.9G)
12:17:26 Join Openwound [0] (~g@
12:17:53pixelmaor at least misleading
12:19:46 Quit Openwound (Client Quit)
12:20:10 Join Openwound [0] (~g@
12:20:51 Join swiftkick [0] (~swiftkick@
12:21:17swiftkickis there a charging issue on the sansa fuze for rockbox 3.6
12:21:44swiftkickbrand new fuze v1, with fresh install of rockbox v3.6 via windows installer
12:21:53 Join timccc [0] (~tim@
12:22:02swiftkickno music on device, all on sd card. played thru all the 1st battery charge
12:22:33swiftkickand now, device doesnt seem to be charging. turning device on, rockbox splash screen, then powers back off because not enough battery
12:22:45swiftkickconnecting USB cable doesnt seem to do anything either
12:23:07swiftkickwhether connected to charger or computer
12:27:15 Quit Llorean (Ping timeout: 272 seconds)
12:28:49swiftkickno access to device anymore via USB. upon inserting USB cable, windows says: "One of the USB devices connected to this computer has malfunctioned. "
12:28:54 Join robin0800 [0] (
12:31:16 Quit krabador (Quit: Sto andando via)
12:33:13 Quit earcar (Quit: zzzzzz)
12:34:35 Join earcar [0] (
12:35:34 Quit kugel (Ping timeout: 245 seconds)
12:36:23 Quit earcar (Client Quit)
12:37:09JdGordonamiconn: you got a minute?
12:37:27JdGordonyour name is at the top of
12:38:05 Quit Openwound (Quit: Http://www.ZeroIRC.NET Zero IRC Ver 2.9G)
12:45:47 Quit Rob2223 (Ping timeout: 240 seconds)
12:46:21 Join Rob2222 [0] (
12:51:05JdGordoncan someone how knows the 2bit lcd drivers please look at ?
12:51:15JdGordonalso anyone who wants to test it...
12:51:44S_a_i_n_tJdGordon: Can I poke you about the eabi toolchain?
12:52:20JdGordonI'll swap you for the images I've requested :D
12:52:36 Join kugel [0] (~kugel@
12:52:39 Quit kugel (Changing host)
12:52:39 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:52:49S_a_i_n_tFair enough.
12:54:34kugelS_a_i_n_t: you can poke me
12:54:59 Quit kugel (Remote host closed the connection)
12:55:11S_a_i_n_tkugel: You have CygWin arm-elf-eabi tolchain?
12:55:46S_a_i_n_tI don't even want to think about what you could mean otherwise!?!
12:56:08S_a_i_n_t...or, perhaps ;)
12:57:05 Join kugel [0] (~kugel@
12:57:07 Quit kugel (Changing host)
12:57:07 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:00:18kugelwhat's your issue with the toolchain? people successfully built it on cygwin
13:00:56S_a_i_n_tit's a PITA, and the necessary packages for eabi aren't doccumented.
13:01:32S_a_i_n_tI bluffed it last time with the help of someone who's nick escapes me.
13:02:26kugelI don't know of any extra package needed except a host toolchain and patch,wget/curl
13:02:59S_a_i_n_tMPFR, and GMP are two I can think of off the top of my head.
13:03:31S_a_i_n_tI just remember it was a real PITA, one I'd like to avoid if possible.
13:03:33kugeldoesn't download those?
13:04:06kugelthey are needed, but you don't need to install them on a Linux box
13:04:10S_a_i_n_tIf it does, it's a new addition...well, new-ish
13:04:18S_a_i_n_tI had to add them to build it initially
13:04:38***Saving seen data "./dancer.seen"
13:04:56kugelthat's possible. I think it downloads since the official switch
13:05:58S_a_i_n_tI'll try set it up to run again while I'm asleep, but, I'd really like to just c/p it from another source if someone feels like being kind to me.
13:06:33S_a_i_n_tI only have access to a pretty slow machine at them moment.
13:06:50kugeljust executing ought to be enough really
13:07:37kugel(after installing wget and patch)
13:07:47S_a_i_n_tI'm trying to trying to remember the name of the guy that I figured out the additional packages with...Japanese guy, doesn;t come in often, rather broken english.
13:08:00S_a_i_n_tHe did the embedded AA patch
13:08:12JdGordonuuchida iirc
13:08:20S_a_i_n_tthat's the one.
13:08:25S_a_i_n_tthat was driving me nuts
13:16:31 Quit kugel (Quit: Yaaic - Yet another Android IRC client -
13:19:25linuxstbJdGordon: Why did you open a bug for the fact that the backdrop is _by design_ an attribute of the screen, and not the viewports on the screen?
13:20:42 Join wodz [0] (
13:20:59 Quit Thanatos (Ping timeout: 245 seconds)
13:21:26JdGordonbecause the design is wrong
13:21:32JdGordonor has become wrong
13:22:16linuxstbYes, but that's not a bug. Otherwise many feature requests could be argued as bugs that way.
13:22:48JdGordonhow does it really make a difference? the patch is all that anyone should be caring about
13:23:00amiconnThe backdrop being for the whole screen is by design, for performance reasons
13:23:13linuxstbThat's what I'm asking - why did you open a bug report, instead of just posting a patch.
13:23:41JdGordonI would be hugely surprised if there was a 0.1% perf loss from this patch
13:23:50amiconnWhat I want to change/ add is that you can specify whether the backdrop should be used or not, by means of another draw mode bit
13:24:00 Join krabador [0] (
13:24:39JdGordonamiconn: thats oretty much what I;ve done, but part of the viewport which I tink is better than drawmode
13:24:54amiconnThe drawmode *is* part of the viewport
13:26:20 Join funman [0] (~fun@rockbox/developer/funman)
13:26:25 Join Dreamxtreme [0] (~Dreamxtre@
13:26:37amiconnAnd the patch will cost quite a bit of efficiency
13:26:45amiconn(in its current form)
13:27:57linuxstbJdGordon: Regarding the patch itself, I don't like the lcd_using_backdrop() function. Firstly, it's badly named (it should be viewport_using_backdrop()), and secondly, it obfuscates the fact that you're just testing the value of a boolean in the viewport struct.
13:28:04pSiKOhi funman , any news about the sd card ?
13:28:14amiconnAdding an if() to the *pixel() functions ia a bad idea
13:28:35amiconnThat's why the drivers are using those function arrays
13:29:11amiconnOne set of functions is used when a backdrop is present, the other when there isn't
13:29:24JdGordonan if which is hit almost never (maybe it need the [UN]LIKELY() macro) would actually add noticable difference?
13:29:51*JdGordon will do a lcd bench on the mini2g in a few min
13:30:30JdGordonlinuxstb: sure, that can be changed, I wanted to get it working first
13:30:57wodzJdGordon: out of curiosity - can you try to view a few jpgs and bmps on your mini?
13:31:03amiconnThat if() is hit for *every* pixel that's drawn by the respective pixel function
13:31:27amiconnAnd (UN)LIKELY doesn't help here
13:31:44JdGordonwodz: yeah, sure
13:32:08 Join Openwound [0] (~g@
13:32:37 Join hebz0rl [0] (
13:32:40funmanpSiKO: no
13:32:49amiconnMy idea doesn't need that if() at all - but it needs clever extension of the function pointer array to allow switching between both cases quickly (i.e. without rewriting the pointers)
13:33:14wodzI can't test right now if gevaerts change from yesterday fixes greylib on COP problem I am experiencing
13:35:48 Quit jfc (Ping timeout: 255 seconds)
13:37:29JdGordonaccording to the wiki, the mini2g does 166.5FPS unboosted, even if that patch drops that by 1% it should be perfectly acceptable. That would give absolutly no noticable impact anywhere
13:37:59amiconnThis has nothing to do with updating fps. It's about drawing speed
13:38:33amiconnUpdate speed is framebuffer -> lcd; this is app code -> framebuffer
13:39:13JdGordoneven more so no big deal
13:39:18linuxstbJdGordon: The actual numbers aren't important IMO. We should always aim to write things as efficiently as possible, especially in something like the lcd driver. Effectively adding an if() inside a loop isn't good if it can be avoided, which amiconn is saying it can.
13:39:23amiconnEfficiency is always a deal
13:40:28JdGordonsure, if someone were to do it fine, but in the absense of that I dont see why slightly less efficient is such a bad thing
13:40:35amiconnIn fact I already had that functionality added half-way a year ago or so, but ran into some problem (which I don't remember anymore, unfortunately)
13:40:37JdGordonand there is a case of overoptimising, but thats an argument I know I'll lose
13:41:25amiconnEver used rb on an ipod video?
13:41:26wodzthere is not something like overoptimising in embedded world
13:41:49linuxstbJdGordon: This isn't optimising, this is taking care to write something efficiently in the first place.
13:42:03amiconnIf you di (or do) you'll see drawing speed is a real issue, at least on some targets
13:42:43JdGordonI use my video occasionally...
13:43:06JdGordonumm... if the seperate functions are there already then its simply a matter of saping the drawing funcs pointer on lcd_set_viewport
13:44:01 Part funman
13:44:33JdGordon2 line change instead...
13:44:34 Quit krabador (Read error: Connection reset by peer)
13:44:37amiconnIt's not that simple
13:45:41JdGordonsurely it is? :D
13:46:07amiconnIt is an array, and copying all pointers around isn't efficient
13:46:22amiconnProbably better than checking an if() for each pixel, but still
13:46:43JdGordonin lcd_set_viewport()
13:46:53JdGordonisnt that what you were suggesting before?
13:47:04JdGordonjust make the driver think there is no backdrop
13:48:05amiconnNot entirelx
13:48:46amiconnWhat happens if you change this property (be it separate or part of the drawmode) for an existing viewport?
13:49:05JdGordonit will take effect every time you select that viewport
13:49:18JdGordonif you do it after calling lcd_set_viewport() you are doing it wrong
13:49:39amiconnThat's why I want it to be part of the draw mode
13:50:11amiconnIt should be selectable without changing the viewport
13:50:25JdGordonso it gets checked before every draw function?
13:50:57amiconnIt gets set whenever you call lcd_set_drawmode()
13:52:22JdGordonhow is that really any better? if this changes during a viewports life it could cause very wierd issues, especially scrolling
13:52:27amiconnActually I think the problem was the behaviour to default to, because the *wrong* default would mean that all app code has to be adapted immediately
13:52:47amiconnNo, because scrolling has to store the drawmode anyway
13:53:44amiconn(and even has to modify it in some cases)
13:54:13JdGordonnothing apart from the driver should need to be touched for this to be added
13:54:21JdGordondefault is use the image
13:55:22wodzpixelma: Your try to build player sim with my patch applied spots apps/talk.c as an another place where lowlevel SH specific code survived :-)
14:00:55 Join krabador [0] (
14:01:29wodzdoes the playback work in hwcodec sims?
14:01:54pixelmathere won't be sound
14:02:36wodzthx, that's what I expected
14:05:40JdGordonnuts, ok apparently it aint working :/
14:07:01JdGordonI think in the skin renderer though maybe
14:11:39wodzJdGordon: what about jpgs/bmps on mini?
14:11:55 Quit Openwound (Quit: Http://www.ZeroIRC.NET Zero IRC Ver 2.9G)
14:12:01JdGordonyou got any I should look at? or just in general?
14:12:33wodzjust in general - I either got crash or screen corruption
14:13:55wodzwell freeze rather than crash
14:14:50 Join teru [0] (
14:15:37JdGordonthey look crap on the lcd, but no problems
14:16:02wodzwhat revision are you using?
14:16:56JdGordonjust built
14:18:29wodzIf this is not to much could you test revision before gevaerts change from yesterday?
14:18:56JdGordonok, hang on
14:19:01JdGordonwhich change?
14:19:13wodzabout plugins loading
14:22:02teruJdGordon: skin paraser should check viewport dimension. do you think change is correct?
14:22:07JdGordonhang/crash with lcd off
14:22:56JdGordonteru: is that really needed?
14:23:44JdGordonwodz: lcd backlight off I mean.. that fix apparently works
14:23:57wodzJdGordon: cool
14:24:30terui think yes. drawing something in the viewport witch is out of the screen make player crash.
14:25:11wodzI was hit by this bug when I was working on png plugin. I spend a lot of time debuging my code before I found out that this is more general problem
14:34:01linuxstbJdGordon: Didn't the viewport parser check the bounds of the viewports in the past?
14:34:21teruJdGordon: my gigabeat X30 freezes with
14:34:21 Quit wodz (Quit: Leaving)
14:34:24JdGordonI dunno
14:34:36linuxstbThe intention when I implemented viewports was that the apps code was responsibile for ensuring the viewports were within the screen.
14:34:54JdGordonteru: run it through gdb?
14:35:36teruit used to be checked in viewport_parse_viewport() or something like that. it is one reason i think the parser should check.
14:36:28JdGordonamiconn: what about this? It still doesnt use the draw mode, but I dont really see how doing so would make it any better
14:37:02teruJdGordon: sorry, i donno how to run gdb on target.
14:37:09JdGordonon sim
14:38:05JdGordon- for font?!
14:38:09JdGordonthat shouldnt be legal
14:38:49S_a_i_n_tAnyone here willing to test the iLike theme for iPod Video? I'm wanting to see if it's up to "release quality" yet.
14:39:15JdGordonactually, maybe it is
14:43:04 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
14:47:26 Join webguest76 [0] (
14:48:09webguest76which computer gbc emulators accept the .sav save file from rockboy?
14:48:45linuxstbrockboy is a port of GNUboy, so maybe that.
14:52:54teruJdGordon: i think i am able to run sim with gdb. what do you want me to do? even if viewport out of the screen doesn't hurt, i think it is better to check the viewport dimension.
14:53:29JdGordonrun gdb and see where it crashes
14:53:31JdGordonand yeah, commit that check if you want
14:59:07 Quit webguest76 (Quit: CGI:IRC)
15:00:50JdGordonamiconn: linuxstb: I've updated 11665
15:00:54 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:04:40***Saving seen data "./dancer.seen"
15:05:08 Part LinusN
15:06:50 Quit JdGordon (Ping timeout: 240 seconds)
15:08:52CIA-7New commit by 03teru (r28244): check viewport dimension when parsing viewport so that player doesn't clash with bad wps.
15:11:03CIA-7r28244 build result: All green
15:11:45 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
15:16:18 Join Kitr88 [0] (
15:17:18Stummi nice. I will try this today in the evening
15:18:46S_a_i_n_tteru: Why do VP dimensions need checking?
15:18:59S_a_i_n_tan out-of-bounds VP won't crash the player.
15:19:05S_a_i_n_tOr, shouldn't.
15:19:52 Quit Kitar|st (Ping timeout: 264 seconds)
15:21:02teruat least, crashes my player.
15:21:18 Quit Kitr88 (Ping timeout: 264 seconds)
15:22:04S_a_i_n_tweird...I just tried it then, editing the VP dimension with the text editor on my ipod, and I can excede the dimensions of the screen quite easily
15:22:46S_a_i_n_ton a 176px screen I tried displaying a 100x50VP at 0x100 and it worked fine
15:24:36terudid you put a text or image on th viewport? the clash seems to happen because lcd functions tries to draw out of lcd_framebuffer.
15:24:54S_a_i_n_tit had a progressbar in it.
15:25:02S_a_i_n_tso, it even dynamically updates.
15:25:05S_a_i_n_tand, is fine.
15:26:19 Join Kitar|st [0] (
15:31:31 Join kugel [0] (~kugel@
15:31:33 Quit kugel (Changing host)
15:31:33 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:32:45 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
15:33:47teruS_a_i_n_t: i don't know why the result is different then. there is still no issue if you put the vp at e.g 0x500?
15:34:02S_a_i_n_tI'll try that in a second.
15:41:25 Quit domonoky (Read error: Connection reset by peer)
15:42:35 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
15:45:26 Join jgarvey [0] (
15:45:34S_a_i_n_tAny iPod Video testers present?
15:57:22linuxstbS_a_i_n_t: The viewport params need checking because passing valid viewports to the LCD driver is the responsibility of apps/ code. The viewport parsing code used to do that, but JdGordon (I assume) removed it.
15:58:34kugelit doesn't use viewport_parse_viewport anymore
15:59:07S_a_i_n_tI think it still matters if the x,y positioning is out of bounds, but it never used to matter if the viewport extended out of bounds.
16:03:40 Join Strife89TX [0] (~cstrife89@
16:04:13linuxstbS_a_i_n_t: Well, Rockbox should reject such viewport definitions as invalid, and IIRC it used to.
16:05:06*S_a_i_n_t has seen plenty of themes with incorrect viewport defs that still load.
16:05:34linuxstbS_a_i_n_t: Recently, or when viewports were first implemented in the WPS?
16:06:09S_a_i_n_tfairly recently, so, unless it suddenly started breaking things in the past few months or still should work.
16:06:23 Quit kugel (Read error: Connection reset by peer)
16:08:39linuxstbYou say those themes load, but do they work reliably without crashing Rockbox? But that would depend on how wrong they were.
16:09:47S_a_i_n_tOf the authors I contacted, from the themes I fixed personally...not one of them had an idea that it wasn't advisable to do so.
16:10:05AlexPS_a_i_n_t: This is more a "how it should be done properly" than "it doesn't happen to crash in some cases so let's not care"
16:10:18S_a_i_n_tI even have a theme that relies on the fact that it is possible to "slide" an image onscreen via an animation.
16:10:37 Quit Strife89TX (Ping timeout: 276 seconds)
16:11:11 Join kugel [0] (~kugel@
16:11:13 Quit kugel (Changing host)
16:11:13 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:11:33S_a_i_n_tThe themeeditor will need some work then, as it doesn;t care one bit about out of bounds VPs
16:11:46S_a_i_n_tand I assume there's a lot of "new themers" using it.
16:13:52S_a_i_n_tQuestion: Would it be OK to pop a theme in the Themeing and Customization Forums for testing?
16:14:01 Join Rhcp1253 [0] (
16:14:12AlexPSure, that's what others do
16:15:11S_a_i_n_tAwesome, I was aware others did so...but I wasn't sure if it was frowned upon at all or not.
16:15:31Rhcp1253I'm just wondering, is anyone actively working on the USB driver for the nano 2g?
16:15:42Rhcp1253Or currently rather
16:15:52S_a_i_n_tRhcp1253: How ddo you mean?
16:16:00S_a_i_n_twhat's "wrong" with it in particular?
16:16:25Rhcp1253It doesn't have full control, as in no docking x
16:16:44S_a_i_n_thow do you mean "no docking"?
16:16:51 Join Strife89TX [0] (~cstrife89@
16:16:51S_a_i_n_tas in, it doesn't mount?
16:17:06Rhcp1253When put on a dock, it resets into disk mode
16:17:25AlexPRhcp1253: Hold menu
16:17:31AlexPIt'll then charge instead
16:17:33S_a_i_n_tthat will depend entirely on the dock you use I would say.
16:17:40AlexPI think it is menu, check the manual
16:17:45S_a_i_n_tSome people are using it with docks just fine.
16:18:01Rhcp1253Is there anyway to get it to play music on the dock?
16:18:03S_a_i_n_tyou probably need to turn on accessory power supply
16:18:13S_a_i_n_tsome docks need to detect accessory power.
16:18:30Rhcp1253Yea, I have that enabled
16:18:52S_a_i_n_tthen, your dock isn;t supported.
16:18:53AlexPRhcp1253: Hold menu when plugging it in so it doesn't go to diskmode, then try
16:19:06S_a_i_n_tthere's very few accessories known to work with the iPods really.
16:19:09AlexPIncidently, this is nothing to do zith USB
16:19:36Rhcp1253Ok, so it's my dock/configuration that's screwy?
16:19:49S_a_i_n_ttry holding Menu while docking it.
16:19:56S_a_i_n_tfailing that, yes.
16:20:01AlexPNo, Rockbox has incomplete support for the apple accessory protocol
16:20:14AlexPOr maybe it is the serial driver, I forget which
16:20:22Rhcp1253Ok, I'll try that an get back to you if I can. Thanks :)
16:21:01S_a_i_n_tNano2G USB is actually pretty solid, it's the HID that's completely borked.
16:21:53 Join leavittx_ [0] (~lev@
16:21:58Rhcp1253Well, a followup question: psgroove has been ported to other rockbox'd iPods, but why not the 2nd gem nano?
16:22:29S_a_i_n_tBecause noone has done the work yet.
16:22:52Rhcp1253So it's just a matter of taking the time to
16:23:00 Join beomuex [0] (~beowulf@unixboard/users/beomuex)
16:23:02Rhcp1253Ok, that makes sense
16:23:09S_a_i_n_ttime/effort...people work on what they want to.
16:23:32Rhcp1253Seems usual nowadays
16:24:27linuxstbRhcp1253: That's how Rockbox has been developed for the last 9 years...
16:25:15 Quit Strife89TX (Quit: Leaving)
16:25:49Rhcp1253Well somethings going right then :)
16:26:05gevaertsRhcp1253: and anyway, psgroove isn't our software. They presumably have their own places to discuss it
16:27:11Rhcp1253Ok, I just figured you guys would have better info on the drivers, which u did
16:27:34 Nick Rhcp1253 is now known as rhcp1253 (
16:29:28linuxstbrhcp1253: All the other ipods use a different CPU and different USB driver. The hardware of the 2nd gen Nano has very little in common with the other ipods supported by Rockbox.
16:29:52beomuexIs there another page besides with all possible key/values for the config.cfg?
16:30:17S_a_i_n_tbeomuex: All should be present there.
16:30:30rhcp1253That's probably why the exploit and development took so long
16:30:47beomuexsure? I cannot find the key that enables the .talk files
16:31:17S_a_i_n_twhy are you looking in the config?
16:31:25S_a_i_n_tall you need to do is set it in the menu
16:31:28beomuexbecause my display is broken ;)
16:31:45linuxstbDon't your menus talk?
16:31:59S_a_i_n_treset your config, it is enabled by default.
16:32:00 Join DSStrife89 [0] (~Strife89@
16:32:16S_a_i_n_tyou may not actually have the voice files present.
16:32:20 Quit piotrekm (Quit: piotrekm)
16:32:26beomuexyeah. They do. But the database entries aren't. The are spelled letter by letter.
16:32:55S_a_i_n_tyou need to create .talk files for them
16:33:07S_a_i_n_tthey aren't handled the same way.
16:33:47beomuexalready done. This page
16:33:57beomuextalks about an option that enables them
16:34:36S_a_i_n_twell, if the menus are voiced, using the manual it shouldn;t be hard to find the option on the device.
16:35:13rhcp1253I g2g
16:35:26rhcp1253Thank you for the help :)
16:35:31 Part rhcp1253
16:42:33beomuexS_a_i_n_t, yeah sure. But this page does not provide a key that has anything to do with the option mentioned in the manual. Deleting the config.cfg enables the spoken output of the main menu but does not enable file spelling (or even spoken filenames using .talk files)
16:44:24 Quit teru (Quit: Quit)
16:44:55S_a_i_n_tbeomuex: the only ones I know of are:
16:44:57S_a_i_n_ttalk dir clip: off
16:44:58S_a_i_n_ttalk dir: off
16:44:58S_a_i_n_ttalk file clip: off
16:44:58DBUGEnqueued KICK S_a_i_n_t
16:44:58S_a_i_n_ttalk file: off
16:44:58S_a_i_n_ttalk filetype: off
16:44:58***Alert Mode level 1
16:44:58S_a_i_n_ttalk menu: off
16:45:33beomuexmany thanks!
16:46:05S_a_i_n_tobviously, you'll want to substitute off with on
16:49:10 Part Zagor
16:49:45 Join jfc [0] (
16:49:48 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
16:54:46 Join pyro_maniac [0] (
16:54:59***Alert Mode OFF
16:57:54 Quit DSStrife89 (Ping timeout: 240 seconds)
16:59:12 Join Judas_PhD [0] (
17:00:01 Quit kugel (Ping timeout: 260 seconds)
17:00:57pyro_maniacthere seems to be an unneeded '>' in line 71. can somebody check this please?
17:02:05CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:02:05*linuxstb wonders what he missed - line 71 of what?
17:03:14pyro_maniacmissed to say the file name :-S me was talking about wps/cabbiev2.176x220x16.wps
17:04:43***Saving seen data "./dancer.seen"
17:05:29 Part Zagor
17:08:13linuxstbpyro_maniac: Yes, that looks wrong to me, but I know nothing about wps syntax.... JdGordon committed that change.
17:09:10pyro_maniaci spotted it by using my e250 and was wondering why this > is there
17:09:26pixelmait's wrong and weird that the WPS isn't rejected
17:09:51linuxstbpixelma: Wouldn't it just be interpreted as text?
17:10:14pyro_maniacseems so
17:10:26pyro_maniacbecause its shown as text
17:10:53linuxstbBut I mean, that's what I would expect to happen, unless ">" is supposed to be an illegal character in WPS text.
17:11:56pixelma">" closes conditional statements, I thought all special chars are interpreted as special, no matter where they are
17:12:08pyro_maniacbut its the only wps with a > after Next Track:
17:12:30pyro_maniacwithout an < before i mean
17:12:35pixelmaotherwise you wouldn't need to escape them with % everywhere, just in some cases, where it could be confusing
17:13:17pixelmathere was no start of a conditional on that line though
17:14:10S_a_i_n_ta < or a > will just be treated as text unless there's a tag prepending the <
17:14:42S_a_i_n_tThat one was probably my fault, strange it took so long for someone to notice it, but, thanks.
17:15:27linuxstbpixelma: I think that's a pretty standard behaviour for language parsers. i.e. the ">" only needs to be escaped when it is ambiguous.
17:16:37pixelmawell, I don't care much and will go on escaping these characters everywhere
17:17:07 Join _s1gma [0] (~d.d.derp@
17:18:08S_a_i_n_tseems unnecessary...I'm fairly certain it needn't be escaped but I'll need to check that.
17:19:05S_a_i_n_tHmmm...apparently it *does* need escaping.
17:19:24*S_a_i_n_t looks at the code to see why it's being displayed.
17:19:48pixelmain this case it shouldn't be there at all
17:21:01S_a_i_n_tNo, that's quite correct.
17:21:12S_a_i_n_tIt *should* choke the parser though.
17:21:21S_a_i_n_tweird it doesn't.
17:22:08linuxstbS_a_i_n_t: What makes you say that it does need escaping. If the ">" in that line is being displayed, then it would seem to me that it doesn't.
17:22:13pixelmait was just my curiousity why the WPS is not rejected, so a single ">" somewhere seems to work. For my personal use, I'll go on escaping it as then it's a "set and forget" even if I later decide to add some conditionals to the line and it could become ambiguous (just a personal opinion)
17:22:45S_a_i_n_tright down the bottom.
17:23:22S_a_i_n_tI have also concluded it wasn't me that fucked it up, so, I'm happy.
17:23:36linuxstbS_a_i_n_t: You mean the list of "Other Tags" ? That just says that a %> displays a ">", not that you _must_ use it.
17:23:42S_a_i_n_tUnless some bastard went and added the crazy stype for subimages.
17:24:00S_a_i_n_tit says it needs to be escaped.
17:24:13S_a_i_n_tor, that is the intention.
17:24:22linuxstbWhere does it say that?
17:24:27S_a_i_n_tit has been this way since the parser (new one) was implemented
17:24:56S_a_i_n_twell, by saying it needs to be prepended with a %, its saying "it won;t work otherwise"
17:25:09S_a_i_n_totherwise, it would be irrelevant to say so.
17:26:21S_a_i_n_tall the chars used as part of arguments are *supposed* to need escaping to be displayed as text.
17:26:23linuxstbI wouldn't read it like that. Reading it literally, it says that "%>" produces a ">", no more and no less.
17:26:31S_a_i_n_totherwise an argument is/should be expected
17:26:56S_a_i_n_tlinuxstb: yes, but you say that without following the development of the parser
17:28:25linuxstbI thought we were talking about what the manual says?
17:28:35S_a_i_n_tif %> produces a > and > produces a >, then what is the point of even mentioning it? surely you must read into it that the intent is to say that "this is how you display a > or a < in case you're wondering"
17:28:39S_a_i_n_tand yes, we are.
17:29:30S_a_i_n_tall the chars that are used in arguments for the skin syntax are supposed to need escaping, otherwise an argument is expected.
17:29:36linuxstbNo, because in other computer languages, the need to escape something can depend on the context. i.e. escaping is not always compulsory.
17:29:42domonokythe difference is that %> always produces a ">", while a > might not do it, depending on the circumstances.
17:30:04S_a_i_n_t">" shouldn;t do it under any circumstances.
17:30:10S_a_i_n_tthe parser should reject it.
17:30:19S_a_i_n_tit is very interesting that it doesn't.
17:30:51domonokywho says that it shouldnt ? that depends on the WPS language definition..
17:31:34pixelmaif I had known that it causes such a discussion, I wouldn't have askes :\ I can follow what linuxstb is saying.
17:31:34linuxstbS_a_i_n_t: I don't feel strongly about that - but if that's the intention, the manual should say something like "The following characters are reserved and MUST be escaped if you want them to appear literally on the screen:"
17:31:54pixelmaor asked
17:32:08S_a_i_n_tthe parser says it shouldn't. if it sees a < it should expect anything before it to be a tag, and complain if there isn't one, and if it sees a > it should be looking for the arguments it is supposed to conatin and complain if there aren;t any.
17:32:42evilnick_BSo which is wrong? The parser or the manual?
17:32:53linuxstbpixelma: I think I just wanted to say that escaping characters is not always compulsory, so it didn't surprise me that the WPS parser didn't object to the > in that line. I have no strong view about whether that's right or wrong in the case of Rockbox.
17:32:55domonokyor both ? :-)
17:32:58S_a_i_n_tthat's why I find it very odd it needn't be escaped, something has changed...that somehow allows it.
17:33:11evilnick_Bdomonoky: Hehe!
17:34:07pixelmalinuxstb: yes, as I said I understand that and accept it
17:34:28S_a_i_n_tthere's even one thing there that's slightly wrong says # displays a # when it should need escaping also.
17:34:41S_a_i_n_tthat's the comment char and should always be ignored.
17:34:58domonokytechnicallly its probably because its a closing tag, and the parser didnt see a opening tag before, so its not in the code path which checks for arguments.
17:35:07S_a_i_n_tI suspect that's atypo though.
17:36:00beomuexS_a_i_n_t, my Sansa 280e Shuffle is working now, thanks
17:36:35S_a_i_n_tNo worries, I'll make a note that those config options aren't in the wiki
17:36:44S_a_i_n_tthanks for pointing it out.
17:38:26beomuexAfter you posted the vars, I found this:
17:38:36beomuexThis should cover it ;)
17:39:05 Join n1s [0] (
17:39:12 Quit n1s (Changing host)
17:39:12 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:40:18*S_a_i_n_t points Nano/Video owners to
17:47:23 Join drizztbsd_ [0] (~quassel@unaffiliated/drizztbsd)
17:49:32 Quit petur (Quit: later)
17:51:02 Quit drizztbsd (Read error: Connection reset by peer)
17:55:44 Quit domonoky (Read error: Connection reset by peer)
17:59:49*linuxstb looks at the skin parser and sees "colour" was used everywhere instead of "color".
18:00:01 Join [sko] [0] (~sko]
18:00:07S_a_i_n_tcolour, is correct.
18:00:17S_a_i_n_tbrittish spelling my boy.
18:00:40linuxstbNo, "colour" is correct for the manual and for comments. Code uses "color" - see the lcd driver for example.
18:01:08 Nick drizztbsd_ is now known as drizztbsd (~quassel@unaffiliated/drizztbsd)
18:01:11linuxstbSo we have things like fgcolour in the skin code, and fgcolor in the lcd code.
18:01:13S_a_i_n_tI thought *everything* was supposed to use UK english?
18:01:29 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:01:33S_a_i_n_tI have seen comments made in the tracker to fix such conventions before now.
18:01:54linuxstbThen they were either wrong, or should have changed the lcd drivers as well.
18:02:13n1snope, only user visible stuff
18:02:15linuxstbBut are you sure they were patches to code, and not the manual?
18:02:35S_a_i_n_tpretty sure they were code patches.
18:02:40n1scode has, as linuxstb pointed out, traditionally been american spelling but i don't think there's a rule for it and it doesn't really matter imo
18:02:50S_a_i_n_treferring to comments being prefferred in UK english.
18:04:32*linuxstb doesn't want to argue about it - the fact is that code has always been US English, and now it's a messy mixture.
18:05:14S_a_i_n_tfix it? whether or not it is UK or US I'm sure it should be consistent.
18:05:34 Join Code-Zombie [0] (
18:05:36Code-ZombieHello kids
18:11:18n1sI think FS #11609 can be closed with a invalid/out of date since the problem seems to be a bios bug that we don't really want to work around and the poster hasn't responded for over a month
18:11:42Code-Zombieis there any hack firmware i can run on my Nano 3g?
18:12:13Code-ZombieGuess ill have to look older :/
18:12:37S_a_i_n_tLooking older won't help, getting an older DAP might though ;)
18:12:40Code-Zombiei have an old Sansa
18:12:45Code-Zombietoo bad its completely broken
18:12:59 Join Strife89TX [0] (~cstrife89@
18:14:39 Join sudoman [0] (92735ab9@gateway/web/freenode/ip.
18:19:15 Join toffe82 [0] (
18:22:56 Quit Strife89TX (Quit: Food get!)
18:27:15alexbobPdoes anybody technical know what would need to be done for rockbox on a nano 3g to happen?
18:27:51alexbobPI wanna know if it's just because nobody's gotten to it, or if I should be blaming apple for locking down their hardware
18:27:57S_a_i_n_tyes, "a lot of work"
18:27:59*alexbobP likes blaming things on apple
18:28:09alexbobPS_a_i_n_t: did apple put roadblocks in the way?
18:28:16alexbobPor is it just yet another port that would have to happen
18:28:46S_a_i_n_tsure, but, they did the same for Nano2G's not like they just leave the devices wide open.
18:29:10linuxstbalexbobP: I think the roadblocks have been mainly overcome. But there is still a lot of hard work to do to understand the hardware, write Rockbox drivers etc. And AFAIK, no-one is actively working on it.
18:29:22alexbobPS_a_i_n_t: well they *should* leave their devices wide open, those jerks ;)
18:29:32alexbobPlinuxstb: okay, I see
18:29:44soapEmail Steve Jobs and ask him why they don't.
18:29:49S_a_i_n_tUm...yeah...think of all the money they'd make then...
18:29:58alexbobPyeah, selling products and all that
18:30:00alexbobPmust be a pain
18:30:02*S_a_i_n_t looks a "this guy's crazy look" ;)
18:30:16AlexPon topic please chaps
18:30:38pixelmalinuxstb: are there unnecessary duplicated now in the code due to "color" and "colour"?
18:30:48alexbobPI was curious because I have a roommate with a 3g... I wanna indoctrinate him into the rockbox fold
18:30:51pixelmaduplicates even
18:30:53alexbobPI was smart and bought a sansa fuze
18:34:19linuxstbpixelma: No, I don't think so. It's just a lot of new code uses "colour", which jumped out at me as unexpected when I read it.
18:39:12 Join panni_ [0] (
18:39:20 Join Maggux [0] (
18:40:09pixelmaguess it increases the chance to get unneccessary duplicates and should be fixed. Now to get someone to do it... :)
18:46:41 Join Strife89TX [0] (~cstrife89@
18:47:56 Join [sko]1 [0] (~sko]
18:48:34 Quit [sko] (Ping timeout: 276 seconds)
18:48:49*linuxstb guesses no-one feels that strongly either way, so assumes changing "colour" to "color", to be consistent with the original decision is the direction to go?
18:49:41S_a_i_n_tprobably, yes.
18:50:13S_a_i_n_tEven though my heart tells me "color" is wrong, if the majority is "color", then so be it.
18:50:40S_a_i_n_tgrep for both, and see which one wins and run with that ;)
18:52:52pixelmaconvention is convention.... which is American English for code, British for UI and explanations in the manual
18:53:11pixelmaas far as I know
18:53:25linuxstbExactly. We could of course change that convention, but does anyone want to?
18:53:26soapexcept in the case of Proper Nouns, where "iPod Color" always wins.
18:53:40soapelse pixelma describes the convention as I recall it.
18:53:42pixelmasoap: it doesn't
18:53:46*linuxstb thought Ipod Color won, or did that get changed?
18:54:21pixelmalinuxstb: if you ask me, yes but it seems nothing won in that discussion yet
18:55:11linuxstbBut the manual is still using Ipod (and Iriver) isn't it?
18:55:32 Join pamaury [0] (
18:55:33 Quit pamaury (Changing host)
18:55:33 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
18:55:50*linuxstb greps and finds both iPod and Ipod, so that obviously wasn't won...
18:55:50soapsorry - yes I typed "iPod" out of habit. I didn't mean to emphasize that, rather "Color".
18:57:04 Quit Strife89TX (Quit: Strife89TX)
18:57:05pixelmamy bet is it's using both somewhere, the platform files which take care of most references to the players have "Ipod" though
19:02:32 Join Llorean [0] (
19:02:40 Quit Llorean (Changing host)
19:02:40 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
19:04:45***Saving seen data "./dancer.seen"
19:06:31 Quit beomuex (Quit: Lost terminal)
19:08:15 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
19:13:57 Quit sudoman (Quit: Page closed)
19:14:45*linuxstb browses the android/ directory and sees that svn:keywords was forgotten
19:15:34pixelma"forgotten"... probably git-svn users
19:17:12 Quit DerPapst (Quit: Leaving.)
19:19:15 Join bertrik [0] (
19:19:16 Quit bertrik (Changing host)
19:19:16 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:19:25*linuxstb wonders what is for, as android/README doesn't say to run it
19:23:53 Join Horscht [0] (
19:23:53 Quit Horscht (Changing host)
19:23:53 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:26:33 Quit robin0800 (Read error: Connection reset by peer)
19:28:18 Join pyro_maniac1 [0] (
19:30:50 Quit pyro_maniac (Ping timeout: 240 seconds)
19:38:06 Part Code-Zombie ("Leaving")
19:40:16 Quit [sko]1 (Remote host closed the connection)
20:03:28 Join perrikwp [0] (
20:11:18 Quit S_a_i_n_t (Quit: I'm only going to Heaven if it tastes like caramel...)
20:14:34 Quit perrikwp (Quit: CGI:IRC 0.5.9 (2006/06/06))
20:14:39 Join webguest14 [0] (
20:14:47 Join perrikwp [0] (
20:15:29webguest14Hey folks, question. I amplugging my iPod 5.5g with rockbox in to my office PC and it is not showing in explorer. Any ideas why?
20:15:48webguest14it works fine everywhere else
20:16:02gevaertswebguest14: do other USB storage devices work there?
20:16:13gevaertsSome office environments explicitely block these things
20:16:29webguest14I just plugged in a USB thumb drive and it pickd it up
20:16:34 Join S_a_i_n_t [0] (S_a_i_n_t@
20:16:46webguest14gevaerts, I am admin, do you know what is blocking it? the setting
20:16:51 Join DerPapst [0] (
20:17:02gevaertsno. I just know this happens in some places
20:17:10webguest14lol thanks
20:17:10gevaertsIf you're admin on that system, that's unlikely I think
20:17:31gevaertsWhich version of rockbox are you running?
20:17:38webguest14I should know considering I am the one who images the machines from scratch and does the initial setup, I cannot think of anything on my PC that should be blocking it
20:18:06gevaertsyes, that's fairly conclusive :)
20:18:28n1swebguest14: that build didn't have usb enabled in rockbox, upgrade
20:18:47gevaertswebguest14: does it show anything on screen when you connect?
20:18:55gevaertsn1s: true, but 3.6 also doesn't have it
20:18:57webguest14Strange as it works like any other usb storage device on every other machine I have ever been on
20:19:07gevaertsAnd it should still reboot to in-ROM disk mode
20:19:09webguest14on the ipod or PC screen?
20:19:16n1sgevaerts: right, i keep forgetting that...
20:19:23gevaertsthe ipod
20:20:00webguest14Yeah it shows the usb and then goes to the screen that says do not disconnect
20:20:09 Join [sko] [0] (~sko]
20:20:23webguest14my PC also makes the usb connected sound but I cannot find the iPod in explorer
20:20:33gevaertsright. That means the ipod is behaving normally.
20:20:52gevaertsThat's actually Apple's in-ROM disk mode, so if it doesn't work, we didn't break it :)
20:21:00gevaertsDo you have itunes on the PC?
20:21:18webguest14absolutley not (I mean no)
20:21:46gevaertsI don't really know then...
20:21:49webguest14I am pretty sure it is an issue on my PC's end
20:22:04 Quit Dreamxtreme (Quit: There are two major products that come out of Berkeley: LSD and UNIX. We don't believe this to be a coincidence.)
20:22:11webguest14I am going to go have lunch with one of my desktop engineers and see what he thinks, thank you very much for your help
20:22:17 Join Dreamxtreme [0] (~Dre@
20:22:22gevaertssounds like a plan :)
20:22:43webguest14Thanks again, goodbye.
20:25:04 Quit Dreamxtreme (Client Quit)
20:25:14 Join Dreamxtreme [0] (~Dre@
20:26:26swiftkickare there fuze v1 charging issues
20:26:55swiftkickdoes rockbox for fuze v1 3.6 require botting in sansa mode to get it to charge
20:27:45swiftkickmine seems to
20:28:11webguest14FYI my desktop engineer had no idea. It worked on his PC (win7) and an XP test machine that is very close to my PC.
20:28:17webguest14Very strange.
20:28:39swiftkickwebguest14: good luck
20:29:09n1sswiftkick: dunno, but a current build on my fuzev1 charges fine in rockbox
20:30:11linuxstbwebguest14: You could try different USB ports, just for fun. Sometimes front-mounted USB ports are less reliable. You could also (if you haven't already) try starting Apple's main firmware on your ipod, and plug it in when that's running.
20:30:44webguest14I have tried 3 ports. Won't apples main firmware want to install itunes?
20:31:23swiftkicknls: how do you know its charging?
20:31:27webguest14strange. It is showing up in the 'safely remove hardware" thing under iPod(no drive letter) but not in explorer
20:31:39linuxstbwebguest14: No (to your itunes question)
20:31:59linuxstbPeople happily use ipods with Windows without ever installing itunes.
20:32:03swiftkickn1s: how do you know its charging besides the fact that the battery size indicator moves?
20:32:13webguest14how do I start it in apples firmware?
20:32:28linuxstbTurn the hold switch on immediately after you turn it on.
20:32:32swiftkickn1s: does the battery indicator blink as well when its charging?
20:32:52gevaertsswiftkick: the debug screen may have battery charging info. I don't know if the fuze has this, but some targets show charging current there
20:33:03 Quit [sko] (Quit: Leaving.)
20:33:16swiftkickgevaerts: i was looking there, found something about battery
20:33:23webguest14All I want to do is transfer my library from the iPod to my PC. Will I be able to from the Apple firmware?
20:33:42swiftkickgevaerts: the problem is, i think, that i used the device till it ran out completely, and so thus did not have enough juice to turn on long enough to get to charging mode (!)
20:34:13n1sswiftkick: no it doesn't blink, the empty space in the battery inicator is animated with a grey thing filling the space, it's very hard to see with the default theme, but i mainly know it works because the battery was almost empty when i plugged it in and it was full when i unplugged it and worked for hours today :)
20:34:25gevaertswebguest14: yes. The Apple firmware does mass storage like anything else, at least provided you don't install itunes :)
20:35:00webguest14it is in the apple firmware and still is not showing up. So strange
20:36:48swiftkickn1s , gevaerts: last night, before booting into the sansa firmware, i ran the device - for the first time since purchasing it - all the way down to the end of the battery.
20:36:48swiftkickn1s , gevaerts: the battery ran low enough that, turning the device on, caused it to shut down immediately, cycling booting up then powering off then rebooting, until i held power switch off.
20:36:48swiftkickn1s , gevaerts: so I tried pluging in my car charger (=12v usb power supply with a standard sansa usb cable)
20:36:48DBUGEnqueued KICK swiftkick
20:36:48swiftkickn1s , gevaerts: it produced the same effect. it turned the device on from an off state, but then endless reboot cycle.
20:36:55linuxstbwebguest14: Has itunes ever been installed on that PC?
20:37:29linuxstbwebguest14: You could check the running processes for "ipodservice.exe" or something similar, which may block the ipod from appearing normally.
20:37:40swiftkickn1s, gevaerts: booting to sansa firmware and charging there, resolved the issue
20:38:07 Quit TheSeven (Ping timeout: 272 seconds)
20:38:27n1sswiftkick: are you using 3.6 release or a current build?
20:38:46swiftkickn1s, gevaerts: once home i plugged it in my home computers usb, still same reboot cycle. i was able to interrupt the rebooting by hitting the down direction button and taking it off the default menu option, to either files, or to settings, but still no charging even after 2 hrs
20:39:02swiftkickn1s: 3.6, i think, whatever rockbock utility 1.2.8 installs out of the box
20:39:18swiftkickversion on the device shows 3.6
20:40:45n1sswiftkick: i would recomend trying a current build then as that works fine for me and it has fixes that were not in 3.6
20:42:09swiftkickn1s: thank you
20:42:50 Join user890104 [0] (
20:43:05swiftkickis 3.7 out?
20:44:07swiftkicknot sure if i saw this in the faq since i didnt think about running rockboxutility again, but rockboxutility does require booting the device in sansa mode yea?
20:44:26n1sno, it shouldn't
20:44:59n1sit just needs a working usb mass storage connection to install a new build
20:45:11n1sso either OF or rockbox usb should work
20:45:23swiftkickn1s: thanks, my last experience was with e200 series with no native usb support
20:48:29Stummiwhen i want to update my rockbox (for some updates in the usb-stack) do i only have to apply the new firmware do i have to update some stuff in the .rockbox-dir too?
20:50:26bertrikinstalling a new rockbox version means you extract a new set of files into the .rockbox directory
20:50:56bertrikupdating the bootloader is usually not necessary
20:50:58n1sStummi: usually you should update everything
20:51:17n1syeah, except bootloader :)
20:52:09 Join Thanatos [0] (
20:52:20 Join MethoS- [0] (~clemens@
20:52:32Stummii want to test this patch:
20:59:47linuxstbStummi: What's your question? Are you asking how to use patches?
21:00:41Stummii am asking if i have to update the bootloader
21:01:30Stummior just copy the .rockbox-files
21:01:47linuxstbMy guess is probably not. Or at least, you could first test with only updating Rockbox itself.
21:01:55Stummii am thinking, since the patch changes the usb-stack i have to update the bootloader, right?
21:04:47***Saving seen data "./dancer.seen"
21:06:01Stummihm, ok
21:06:01linuxstbI don't know if the usb stack (or any usb code) is done in the bootloader. But if a bootloader update was needed, that task should have said so - the assumption is generally that it is not.
21:06:01linuxstbBut you could add a comment to that task to ask - it's a fair question.
21:06:01 Quit webguest14 (Quit: CGI:IRC (Ping timeout))
21:06:01 Quit perrikwp (Ping timeout: 245 seconds)
21:06:01swiftkickn1s or anyone: so the place to get the latest build for my player is ?
21:06:01CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:06:01*gevaerts doesn't really see how that patch could be related to the bootloader
21:06:01Stummilinuxstb, i think, the decission to boot into the of or rb is done by the bootloader, not?
21:06:15 Join HypothesisFrog [0] (
21:06:20linuxstbStummi: Yes, but that could be a simple GPIO pin check - i.e. it may not touch the usb hardware in any way, and hence won't be affected by this patch.
21:06:33HypothesisFroghi. Are there any 16GB+ mp3 players supported by RockBox?
21:06:50 Join wodz [0] (
21:06:56wodzpixelma: ping
21:07:11linuxstbHypothesisFrog: Many. See the list of supported devices on the frontpage of the Rockbox website.
21:07:24gevaertsStummi: the requested tests aren't about booting with USB plugged in. They're about plugging USB in
21:07:36HypothesisFrogI mean I'm looking at this page, and it shows a heap of devices, but doesn't list what their capabilities are.
21:07:45HypothesisFroglinuxstb already looking at it.
21:08:52swiftkickn1s: ok this is weird. actually the problem seems to center on the USB connector. fully charged under sansa firmware, powered down. plugged cable in and it did the reboot cycle thing again. computer says "a usb device has malfunctioned"
21:08:53linuxstbHypothesisFrog: Try - and be aware that very few of those devices (if any) can still be bought new.
21:08:53swiftkickn1s: so i reboot to sansa firmware. device recognized. both drive letters present.
21:09:36HypothesisFroglinuxstb thanks. So rockbox is pretty far behind?
21:10:25linuxstbHypothesisFrog: Porting Rockbox to a device is a lot of hard work - by the time it is done, the device is often discontinued.
21:10:48linuxstbmp3 players have a relatively short lifespan before being replaced by newer models.
21:11:09wodzlinuxstb: You failed to mention it is a lot of fun also :-)
21:12:25Stummibut in some cases the newer models doesn't differ much from the old ones and the porting for this isn not so hard ;)
21:13:25HypothesisFroglinuxstb ty
21:13:32pixelmalinuxstb, HypothesisFrog: the DeviceChart lacks a few newer players, e/c200s in ther v2 series, Fuzes, Clips
21:13:36pixelmathe Cowon D2
21:14:00pixelmaalthough the latter still has some problems I think
21:14:43gevaertsOn the D2, the features that are advertised as working work fine. It's just that some features that people consider to be essential aren't advertised :)
21:15:03gevaertsThe main one being that rockbox can't write to internal flash
21:15:06pixelmaall of those don't have 16GB storage internally but most of them (except the v1/v2 Clips), it can be expanded with a microSDHC
21:15:48pixelmagevaerts: didn't it also have intermittent problems since the eabi switch?
21:15:52 Quit saratoga (Ping timeout: 265 seconds)
21:16:04gevaertspixelma: were
21:16:18gevaertsn't those solved yet? Maybe...
21:16:29*gevaerts isn't sure
21:16:48pixelmaI thought I read there are more which come and go with builds (in the new ports thread)
21:17:00pixelmawodz: if you want to ask about testing, sorry I didn't so far
21:18:03pixelmakeep on nagging :)
21:18:11 Join kazaik [0] (
21:18:14 Quit Thanatos (Quit: Leaving)
21:18:46swiftkickwhat is multimedia keypad mode?
21:18:55 Part HypothesisFrog
21:19:39swiftkicki just updated to the latest build and rockbox doesnt reboot endlessly
21:19:53swiftkickbut i have something on my sansa screen that says "USB Keypad Mode"
21:20:42swiftkickthank you gevaerts:
21:21:06swiftkicklol why is this on by default
21:21:14n1swhy not?
21:21:39gevaertsswiftkick: I also just realised what was causing your reboot problem...
21:21:56swiftkickwell, in rockbox mode im still getting 'a device plugged into the computer has malfunctioned' and/or 'a usb device is not recognized' and no drive letters
21:22:11swiftkickgevaerts: /me looks
21:22:24 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
21:22:33gevaertsBasically a mismatch between 3.6 and the latest bootloader on fuze
21:23:01wodzanyone here have some experience with this jtag scanner (
21:23:10*linuxstb wonders if we're feature freezing today
21:23:17gevaertsswiftkick: you could try disabling HID, but with most systems it should work without problems
21:23:26gevaertslinuxstb: good question :)
21:24:11gevaertsAfter last weekend's bugfixes I think we can talk seriously about releasing, so I think I'm in favour
21:24:15swiftkickgevaerts: cool i just upped to the latest build from and it recognized the drives and such after a 2nd reboot
21:24:27bertrikI guess we just need someone saying we are in freeze in a stern voice (and be taken seriously)
21:24:42linuxstbgevaerts: Well, you seem to be the person that's looked in depth at the outstanding bugs, so I'll bow to your judgement.
21:25:07swiftkickgevaerts: the main reason i can think of for not wanting this usb presentation mode is, what about when i plug this in my car usb power supply
21:25:10pixelman1s: actually it should be off by default since it can block normal USB data transfers on some OSs
21:25:30gevaertsswiftkick: if that's just a power supply, it won't matter at all
21:25:34swiftkickis it going to recognize its JUST charging and not go into HID mode?
21:25:53swiftkickgevaerts: excellent, thanks, i am hereby reassured. =)
21:26:00swiftkickmostly because of how helpful everyones been
21:26:02gevaertsit should
21:26:37n1spixelma: isn't that only one OS, macosx 10.4 or something?
21:27:09gevaertsn1s: I believe some other people reported problems too, but nothing really as clear as the OSX 1.4 case
21:27:11pixelmathat one is known and probably older ones. I still think it's important enough
21:27:30swiftkickgaverts: thank you and everyone else who helps keep rockbox runnin, this refurb fuze v1 cost me like, $30 at the local tiger direct, and makes a truly sweet combo with rockbox
21:27:47n1swell, sure, it will be annoying for users of that system
21:27:54swiftkickwas getting worried maybe it was a janky usb interface cuz its refurb
21:28:03pixelmaas data connection is the one you really expect to work with USB on a media player, HID is just an addition
21:28:28 Join webguest14 [0] (
21:28:30webguest14K folks I am back
21:29:12webguest14having issue with 5.5g i pod with rockbox (3.4?) installed and it will not show up in explorer on only this one machine
21:29:16wodzI would say that having recurrence in skin engine (on static allocated systems) is asking for troubles. Jd asked for help in turning this into iteration. I think we should help him before we trigger another "hard to spot (TM)" bug
21:30:01webguest14I am looking in device manager and have found the driver for the ipod under the generic "USB mass storage device" name
21:30:31gevaertswebguest14: that sounds right
21:30:33webguest14when I uninstall and plug hte ipod back in it pops up like the pc detected it but then says "device software installed successfully"
21:30:41webguest14I still cannot see it in explorere
21:30:42linuxstbwebguest14: Has itunes ever been installed on that PC?
21:31:04webguest14linux this will sound wierd but I don't know, but I don't hink so
21:31:09webguest14if it was it was since uninstalled
21:31:33linuxstbYou could also check the task list for any processes running with "ipod" in their name - itunes installs a service which can block disk access.
21:31:38webguest14I learned from my desktop engineer that this machine was most likely NOT re-imaged before I was given it
21:32:09webguest14linux all I see is "ipoint.exe"... nothing else that even starts with "i"
21:32:33webguest14well iexplore//
21:33:23linuxstbwebguest14: OK, it was just an idea... Now I'm out of those.
21:34:24linuxstbBut this is getting off-topic for this channel - as it's happening in Apple's firmware, it's highly unlikely to be something Rockbox is doing. You're welcome in #rockbox-community though (most of the people here are there).
21:35:04webguest14Well thank you for your help.
21:35:07 Quit webguest14 (Quit: CGI:IRC)
21:36:58n1shmm, a thought on releases: one bigish problem seems to be (to me) that we are effectively not able to support release builds as time goes by and the trunk moves forward we (or at least I) don't remember what was broken and has since been fixed or what was disabled etc in the last release and we always ask users to try with a current build when they report bugs
21:37:49n1salso there has been at least a few caases of bugs reported against 3.6 in fs that are really against trunk as well as the already fixed bugs being reported several times in the releases
21:38:30gevaertsI think we're overdoing that though. Asking users to try with a current build before reporting bugs is fine, but sometimes some people tend to point to current builds straight away, seemingly without even looking what the problem is
21:38:44n1si think this is an argument to either release often (the 4 times/year seemed to work fine) or do more point releases
21:39:12wodzbugs reported agains 3.6 is FS thing - you have no possibility to point to svn revision when submitting bugreport
21:39:29gevaertsI think it's mainly an argument for not doing sloppy released
21:39:36n1swodz: sure that is mostly an FS issue
21:40:13wodzI'd love to see possibility to report issue/patch agains particular svn revision
21:41:02n1sgevaerts: we do have the problems with things like the recently mentioned fuze bootloader that doesn't quite work with the release ( i guess that is mostly a rbutil/manual issue though) and the clip being stable, etc, people (read devs) seem to just not think about the last release
21:41:10n1swodz: me too
21:41:53n1sas time goes on, as it has since 3.6, the issues accumulate
21:42:03evilnick_Bn1s: Agreed, the Fuze v2 install instructions are not very clear
21:42:57Lloreanwodz: Isn't there a "current build" option too (or daily perhaps?)
21:43:05LloreanThere is a way to mark a bug as 'newer than the latest release'
21:43:11gevaertsn1s: I don't think we can blame the manual or rbutil for that. We have one released bootloader, and if the stable release and the current; build have different requirements, that's a problem
21:43:49pixelmaLlorean: IIRC there isn't
21:44:19n1sthere is a "daily build" you can choose
21:44:32pixelmaas I said yesterday, I don't remember details just that I had trouble finding something suitable for my bug reports
21:44:37LloreanWe just need to change the text on that to "recent SVN build" or something
21:44:41 Join cardamon [0] (~mute@
21:44:45LloreanAnd then ask people to specify the number in the task description
21:44:53n1sgevaerts: yes but should the bootloader be fixed or the instructions/installer?
21:44:57LloreanI'm pretty sure the "daily build" choice includes "specify"
21:45:02LloreanOr "which"
21:45:18gevaertsn1s: changing instructions or the installer won't help
21:45:59 Join Thanatos [0] (
21:46:09cardamonHmm. I'm still getting used to making playlists from my player itself. There's no way to really quickly add a single file to a playlist rather than the whole directory, is there?
21:46:24gevaertsEither release a bootloader together with the new release, or release the new bootloader later, and live with the fact that it will boot to the OF for a while after we can handle USB
21:46:29n1sanyway, i (now figured out that I) mean that old releases are a support burden for us and an unplesant experience for users so we need to either put some effort into making sure the release works until the next one comes around or make releases frequent enough that it doesn't matter too much
21:46:45AlexPcardamon: I think you can set the hotkey to add to playlist
21:46:45 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
21:46:53AlexPcardamon: Have a shufti at the manual
21:46:55evilnick_Bcardamon: Yes there is. Long-select on the file then Insert/Queue
21:47:11AlexPer, shufti = look :)
21:48:04n1sgevaerts: if the installer downloaded the bootloader that worked for the release and the instructions told users which one to get we would be fine, and not it wasn't so serious in this case but it's an example
21:48:25gevaertsn1s: yes, but the bootloader that worked for the release is not supported :)
21:48:38n1sright :)
21:49:39gevaertsBasically the best way to have good releases is to pretend we don't have releases so the current build has to be stable at all times :)
21:50:11gevaertspixelma: did you have some time to see if the screen corruption issue is still there?
21:50:14n1sthat is the other way, right now we are not doing either of them though :)
21:50:29gevaertsYes. That's what's going wrong
21:51:43gevaertsAnd I sometimes suspect that some people feel they can be a bit less careful because we can point people to 3.6, forgetting that we made life hard for people who stay with 3.6 as well...
21:52:23*gevaerts still isn't really happy about how the theme site handles the theme syntax change
21:53:13linuxstbHow was that handled?
21:53:51linuxstbOuch, there's an "oldsite" linked from
21:54:09 Join webguest14 [0] (
21:54:41gevaertsFrom memory, first 3.6 support was going to be dropped, then it there was going to be a read-only copy, and finally the current oldsite solution was done
21:54:45webguest14Just wanted to let you guys know that I figured it out. For some reason the iPod was naming itself drive H: and I already have a drive H mapped
21:55:03gevaertswebguest14: why didn't it take J: then?
21:55:04webguest14I disconnected that drive and the iPod popped up immedietly
21:55:09AlexPwebguest14: Ha, good work. Thanks for letting us know
21:55:16AlexPSomething to keep in mind
21:55:18webguest14I have no idea why it would not take the next letter
21:55:20gevaertsOh, I think I've seen that too. Windows bug I'd say
21:55:23webguest14like it should have
21:56:18webguest14Yeah went to diskmanagemtn and ipod was listed as drive h. Wierd stuff. Thanks for your help all!
21:56:20 Quit S_a_i_n_t (Quit: I'm only going to Heaven if it tastes like caramel...)
21:56:30*linuxstb wonders if "breaking all themes" is an excuse to go to 4.0
21:58:02 Quit webguest14 (Client Quit)
21:59:02linuxstbgevaerts: What about a unified themes site, where you select the version of Rockbox you're running, and you get compatible themes? Or was that something no-one was motivated to implement.
21:59:25n1swould moving rbutil into its own "project" in flyspray and having a "reported in version" with "Release x.y" and "Current/SVN build" for both rbutil and rockbox?
21:59:38gevaertslinuxstb: I suspect nobody thinks the theme site is very exciting to work on
21:59:38pixelmagevaerts: sorry, no. I admit being a bit lazy and tired of yesterday's testing session. I will try to spend some time the rest of the day and tomorrow
21:59:39n1swould it make things clearer/better, i mean?
21:59:53gevaertspixelma: I can definitely understand that :)
22:01:28gevaertsSpeaking of flyspray, shouldn't we review then "User Interface"/"Themes" distinction? More and more of the UI is themable, so this becomes blurry
22:01:51n1sgevaerts: probably
22:02:32pixelmayeah, I wondered in some cases which category the bugs I reported fell under
22:03:20n1smerging those two into one would probably be a good idea
22:17:08 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
22:19:23linuxstbDo wps files now include a version number?
22:19:43*linuxstb reads an email on the -dev list from Zagor suggesting it, but no-one seemed to reply to it.
22:21:53*linuxstb can't find that much discussion at all about the skin changes on the list
22:21:59 Quit TheSeven (Ping timeout: 272 seconds)
22:22:31gevaertsI think it was mostly here
22:24:55 Quit leavittx_ (Ping timeout: 264 seconds)
22:24:59linuxstbOK. So I'm guessing there's no version number, looking at the CustomWPS page.
22:25:47n1sdon't think there is, no
22:25:53 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
22:27:20 Quit TheSeven (Read error: Connection reset by peer)
22:27:23wodzTheSeven: You have pointed me to jtag scanner - do you know/have the documents describing how it works?
22:27:24 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
22:28:32 Quit n1s (Quit: Lmnar)
22:30:04 Join gourd_captain [0] (
22:30:29gourd_captainI'm having some problems with my newly purchased Sansa Fuze (v2, 8GB) and my microSD card.
22:30:52gourd_captainWhenever I access a file from it, there's a noticeable (20s?) lag before playback.
22:31:25Tornetry turning on the directory cache
22:31:42gevaertsgourd_captain: that's on the microSD card, and not on internal memory?
22:33:03gevaertsSounds like a problem for funman...
22:33:44gevaertsI believe there are some issues with some cards on fuzev2, which some people are looking into. I don't know details though
22:34:12*Torne still suggests trying the directory cache though
22:34:16*gourd_captain had a Clip+. Had being the emphasized word - water submersion does bad things to tech.
22:34:49gevaertsDid you use this same card in the clip+?
22:34:54 Join fdinel [0] (
22:35:07 Join soap_ [0] (~soap@rockbox/staff/soap)
22:35:17gourd_captainDidn't think water would damage it, but who knows.
22:35:45 Quit soap (Disconnected by services)
22:35:47 Nick soap_ is now known as soap (~soap@rockbox/staff/soap)
22:35:49gourd_captainGot this one for a less terrifyingly bad screen.
22:36:06gevaertsSo the card got wet too?
22:36:16gourd_captainFor like, three seconds, tops.
22:36:18gevaertshm, yes. Hard to say then
22:36:40gourd_captainDirectory Cache seems to have improved it immensely.
22:36:45gevaertsIt could be a rockbox issue, or it could just be the card
22:36:45 Quit bmbl (Quit: Bye!)
22:37:02gourd_captainI just didn't turn it on first because it was the source of major, major stability issues on the Clip+ when enabled.
22:37:22Torneit may just be that your sd card is really slow :)
22:37:43Torneor yeah, it might be our problem
22:37:48gourd_captain16 GB Sandisk?
22:37:55Tornesure, lots of sd cards are slow
22:37:57Tornemost, in fact :)
22:37:59gourd_captainFair enough.
22:38:07gourd_captainMost SD cards are crap.
22:38:24gourd_captainI wasn't looking for high transfer rate, just music playback.
22:38:27Tornebut yeah, if it reads reliably but slowly, then it could be the problem gevaerts mentioned
22:38:28gevaerts20 seconds sounds excessive though
22:38:46gourd_captainWith Cache turned on, it works instantaneously.
22:38:48Tornegevaerts: yah, but not impossible
22:38:54gevaertsI'd say don't throw it away. A future update may improve things
22:38:55Tornegevaerts: if the fs is fragmented a lot
22:39:03gourd_captainThis SD card does have 3200 music files loaded onto it.
22:39:09Torneit might be something that we can fix in the long run
22:39:23Tornebut if everything *works* otherwise and it's just slow then turning the dircache on is a reasonable workaround for now :)
22:39:43Torneif we figure something out and the underlying problem gets fixed you can probably turn it back off then :)
22:39:59gourd_captainMemory card is mostly fixed content, and the internal flash is what I use for podcasts.
22:40:27 Join leavittx_ [0] (
22:40:31gourd_captainYeah, on the Clip+ cache would cause it to segfault when I opened the database. Not a good deal.
22:43:26pixelmaif the trouble was the card, couldn't it be tested using the OF?
22:43:54gourd_captainYeah, well, problem there is telling the difference between the card being slow as crap and the OF being as slow as hell.
22:44:04gourd_captainIt doesn't react too well to this volume of files.
22:44:33pixelmaor a card reader maybe if you got such a thing
22:44:45 Quit wodz (Quit: Leaving)
22:44:47gourd_captainOkay - files start instantly in OF.
22:45:01gourd_captainIt shows like three of my files (lack of codecs) but still.
22:45:31gourd_captainPlus side, playing doom on this is actually playable.
22:45:45gourd_captainAs opposed to the hilarious dithering that results on a Clip+.
22:45:58*Torne still doesn't understand why anyone plays doom on rockbox.
22:46:13gevaertsSays the one who ported frotz :)
22:46:39 Join liar_ [0] (
22:46:59gourd_captainI played Slouching Torwards Bedlam on Frotz on the Clip+ once. I was REALLY bored.
22:47:19Torneyeah, you must've been
22:47:29Tornefrotz is less terribleon players with a 4-way pad
22:47:33Torneon the ipod it's just horrendous
22:47:34 Quit liar_ (Client Quit)
22:47:51 Quit gourd_captain (Quit: Leaving)
22:47:54 Quit leavittx_ (Ping timeout: 240 seconds)
22:47:54 Join liar_ [0] (
22:48:42 Quit cardamon (Ping timeout: 264 seconds)
22:48:43Tornein theory i ported frotz so people could use it as a target to write games rather than for people to play classic text adventures
22:48:46Tornebut it works, so hey
22:57:49 Join edboyer93 [0] (
23:00:22 Join leavittx_ [0] (~lev@
23:04:48***Saving seen data "./dancer.seen"
23:13:57 Quit Thanatos (Ping timeout: 245 seconds)
23:15:32 Quit xnyhps (Ping timeout: 276 seconds)
23:19:02 Quit Maggux (Read error: Connection reset by peer)
23:20:20 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
23:21:21 Quit krabador (Quit: Sto andando via)
23:24:19 Join sudoman [0] (97cce263@gateway/web/freenode/ip.
23:26:24 Join xnyhps [0] (
23:35:20 Quit dfkt (Quit: -= SysReset 2.53=- Sic gorgiamus allos subjectatos nunc.)
23:39:47 Quit bertrik (Ping timeout: 276 seconds)
23:40:17 Quit komputes (Read error: Connection reset by peer)
23:44:59 Join perrikwp [0] (
23:49:46 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
23:56:35 Quit ender` (Quit: Shooting is not too good for my enemies. -- Evil Overlord List #4)
23:57:25 Join Thanatos [0] (

Previous day | Next day