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

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

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

#rockbox log for 2014-01-15

00:08:18 Quit ender` (Quit: "HR?" "Human Resources." "In Brussels that kind of department is referred to as the Office for Personkind Enablement. Resources sounds like something you dig out of the ground." -- Peter F. Hamilton: Great North Road)
00:10:56JdGordonkugel: i really dont like the idea of complelty dissalowing text
00:11:28kugelcan you propose an alternative?
00:12:23kugeldisabling themes altogether really sucks, especially when not even this works (themes somehow get re-enabled, I wonder about side effects)
00:13:52 Quit Tomwi (Read error: Connection reset by peer)
00:14:05 Join Tomwi [0] (
00:14:28 Quit Tomwi (Read error: Connection reset by peer)
00:15:21JdGordonthemes are getting reenabled because of the interaction between the viewport_theme_enable() and skins
00:17:01 Quit pamaury (Ping timeout: 260 seconds)
00:17:32 Quit lebellium (Quit: ChatZilla [Firefox 27.0/20140109165205])
00:18:44kugeldisabling still sucks
00:19:06 Join Tomwi [0] (
00:19:34 Quit Tomwi (Read error: Connection reset by peer)
00:19:53 Join Strife89 [0] (
00:23:41 Quit kugel (Ping timeout: 272 seconds)
00:24:06 Join Tomwi [0] (
00:24:29 Quit Tomwi (Read error: Connection reset by peer)
00:29:09 Join Tomwi [0] (
00:29:33 Quit Tomwi (Read error: Connection reset by peer)
00:34:06 Join Tomwi [0] (
00:34:30 Quit Tomwi (Read error: Connection reset by peer)
00:39:10 Join Tomwi [0] (
00:39:34 Quit Tomwi (Read error: Connection reset by peer)
00:40:46[Saint]I hate to be "that guy"…but "I haven't seen an attempt at a better solution in years so I'm willing to accept whatever" doesn't seem like the right attitude.
00:41:09[Saint]Just sayin'.
00:44:10 Join Tomwi [0] (
00:46:25 Quit bertrik (Ping timeout: 260 seconds)
00:52:47[Saint]Would it be difficult to check if we couldn't just drop all preloaded .sbs fonts into RAM?
00:53:20gevaertsCheck, sure. But what then?
00:53:32[Saint]Fall back to sysfont on LOWMEM and cases where font(s) dont fit in RAM.
00:54:02[Saint]Stuff the required fonts in RAM if we can and we're golden, no?
00:54:24***Saving seen data "./dancer.seen"
00:55:30*gevaerts predicts theme people complaining loudly about the theme site offering themes for targets they didn't personally test them on so they look ugly for some people
00:57:41[Saint]It seems a shame to punish all cases simply because a subset of targets doesn't have the RAM available.
00:58:18[Saint]On some targets there's enough to stuff 10+ unifonts
00:59:10[Saint]But sadly, I can imagine the scenario you propose too.
01:01:32JdGordonat the very least we need to figure out what text people would actually want to display
01:01:47JdGordona clock seems reasonable
01:02:02JdGordonand some static lines
01:02:18gevaertsI think this is a matter of how to describe things. Instead of "We haven't had a solution for years, so this will do". let's go for "This seems like a good first step" :)
01:04:07gevaertsI don't think kuge;
01:05:09gevaertsI think kuge's proposal is a step forward, although it can be argued if it's a very big step, and I don't think it would actually hinder future improvement
01:05:37gevaertsAlso, it might wake people up. A years-old status quo tends not to be fresh in people's minds
01:06:31JdGordonwell, his diff also is missing all the needed bits to know if it is in a sbs or not
01:07:01PurlingNayukiExam today :(
01:07:07JdGordongood luck!
01:07:46gevaertsJdGordon: it's on a pastenbin, not on gerrit. It's not supposed to be finished :)
01:09:01JdGordonwe allow the radio to play in usb dont we?
01:09:34gevaertsI'm not sure. We don't stop it, but I'm not sure if that's by design
01:12:42JdGordonno reason to not, but that means people will want to show the fms while in usb
01:12:51JdGordonwhich is reasonable but potentially difficult
01:19:18[Saint]Keeping FM alive in USB seems odd.
01:20:43[Saint]There's no reason to not do so, I just can't think of any other device that offers anything more than a mountpoint when connected via USB.
01:21:09gevaertsIsn't that enough of a reason? :)
01:21:57[Saint]It seems like killing FM playback might make things simpler.
01:22:19[Saint]I wonder how many people even use it.
01:24:07gevaertsKilling it is simpler than handling it properly with controls and things, of course, but that's not what we're doing
01:24:12gevaertsWe just don't kill it
01:24:14 Join cmhobbs [0] (
01:24:14 Quit cmhobbs (Changing host)
01:24:14 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
01:27:59 Join ParkerR [0] (ParkerR@unaffiliated/parkerr)
01:54:00 Quit cmhobbs (Ping timeout: 265 seconds)
01:55:41 Quit ac_ (Remote host closed the connection)
02:00:47 Join krabador [0] (~krabador@unaffiliated/krabador)
02:31:31 Join krabador_ [0] (
02:31:40 Quit krabador_ (Read error: Connection reset by peer)
02:37:08 Quit krabador (Quit: Sto andando via)
02:54:26***Saving seen data "./dancer.seen"
04:08:59 Join Jbob [0] (4b64380c@gateway/web/freenode/ip.
04:19:36 Join cmhobbs [0] (
04:19:36 Quit cmhobbs (Changing host)
04:19:36 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
04:25:33 Quit Jbob (Quit: Page closed)
04:54:21 Quit pixelma (Disconnected by services)
04:54:22 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:54:22 Quit amiconn (Disconnected by services)
04:54:22 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:54:24 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:54:25 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:54:28***Saving seen data "./dancer.seen"
05:03:44 Quit cmhobbs (Ping timeout: 272 seconds)
05:17:31 Join plain-user [0] (
05:38:38 Quit K1773R (Read error: Operation timed out)
05:46:48 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
05:55:42 Quit [7] (Disconnected by services)
05:55:55 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:59:47 Nick SuperBrainAK is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
06:10:56 Quit mt (Ping timeout: 245 seconds)
06:13:12 Quit plain-user (Quit: Leaving)
06:54:32***Saving seen data "./dancer.seen"
07:01:51 Join dfkt_ [0] (OxO29A@unaffiliated/dfkt)
07:03:26 Quit dfkt (Ping timeout: 272 seconds)
07:22:04 Join Rower [0] (
07:42:20 Quit nosa-j (Ping timeout: 272 seconds)
07:44:32 Join nosa-j [0] (~m00k@
08:11:51 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
08:16:53*[Saint] wonders how there is someone viewing a topic that shouldn't ever have been public
08:17:29[Saint](hit the authorization for first posts and bounced straight into the crap can)
08:22:13[Saint]I just noticed in the "who's viewing what" thingo that someone was on a topic that should have been in the shit can and thus invisible.
08:24:09 Join wodz [0] (
08:25:13 Join ender` [0] (
08:36:27 Join einhirn [0] (
08:37:28ParkerR[Saint], fancy seeing you here ;)
08:38:29 Quit nosa-j (Ping timeout: 251 seconds)
08:41:11 Join nosa-j [0] (~m00k@
08:41:58 Join bertrik [0] (
08:41:58 Quit bertrik (Changing host)
08:41:58 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:42:03wodzI am wondering if recent memory corruptions fixed stabilize usb on ams. Has anyone conducted some experiments?
08:49:59 Quit [Saint] (Read error: Connection reset by peer)
08:52:08 Join [Saint] [0] (~saint@rockbox/staff/saint)
08:52:16 Join onder`_ [0] (
08:53:32 Quit onder` (Ping timeout: 276 seconds)
08:53:37 Nick onder`_ is now known as onder` (
08:54:33***Saving seen data "./dancer.seen"
09:05:18 Join kugel [0] (~kugel@rockbox/developer/kugel)
09:08:37 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
09:11:58 Quit wodz (Quit: Leaving)
09:15:18 Join lebellium [0] (
09:18:30kugelJdGordon, gevaerts: we don't stop the fm radio but we don't show the radio screen either
09:18:43kugelonly the SBS can be drawn, the UI viewport is filled with the USB logo
09:19:37lebelliumFM radio on USB screen makes no sense
09:20:06 Quit bertrik (Ping timeout: 245 seconds)
09:20:18kugellebellium: what do you mean?
09:20:36ParkerRI want a Rocjbox player with FM radio
09:20:56ParkerRMy only experience has been an old iPod Mini and a nano 2g
09:22:24lebelliumkugel: oh my bad, I did not notice the radio still works during the USB connection -_-
09:23:45lebelliumbut there are no playback control. Impossible to change the volume or the frequency during USB connection
09:24:24lebelliumso it's a bit strange to get it work during USB connection
09:26:07 Join maruk1 [0] (
09:29:08 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:39:37 Join JdGord [0] (
09:48:38 Quit Rower (Ping timeout: 272 seconds)
09:51:30 Quit nosa-j (Ping timeout: 260 seconds)
09:52:42 Join nosa-j [0] (~m00k@
09:53:34 Join wodz [0] (
10:03:48kugelwhat would be a good placeholder glyph?
10:04:48wodzkugel: Any other remarks then lack of core_check_valid() in buflib crc patch?
10:06:20kugelwodz: I think a core_* wrapper should be created and used in thread-sdl.c
10:06:53wodzYes, I understand. I am asking if the patch is OK otherwise
10:08:55kugellooks good
10:14:44 Quit nosa-j (Ping timeout: 265 seconds)
10:17:01 Join AndChat|694784 [0] (
10:17:14 Join nosa-j [0] (~m00k@
10:19:45 Quit JdGord (Ping timeout: 252 seconds)
10:20:37 Quit AndChat|694784 (Client Quit)
10:20:52JdGordonthere is no particular reason why we couldnt allow full radio control in usb tohuhg
10:21:09gevaerts*nearly full control
10:21:27gevaertsNo saving or loading presets, or finding station art!
10:22:12 Quit pamaury (Ping timeout: 272 seconds)
10:25:37 Quit mc2739 (Read error: Connection reset by peer)
10:26:22 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
10:28:50PurlingNayukiBut basic control such as channel switching, pause, play and stop(start) is possibly.
10:30:45kugelJdGordon: there is a reason: usb hid
10:31:10kugelbut I think hid can be disabled in which case radio control could be granted
10:32:24kugelbut that's a step too far, we're still at the usb_screen, which we need to fix
10:32:43kugeltheme handling at usb screen
10:54:36***Saving seen data "./dancer.seen"
10:57:11 Quit Tomwi (Read error: Connection reset by peer)
10:57:13 Join AnLe[away] [0] (
10:57:28 Join Tomwi [0] (
10:57:35 Nick AnLe[away] is now known as AnLe (
10:58:10AnLecan someone explain me some thing about development builds:
10:58:22wodzsure, ask
10:58:27AnLei wan't backup my 3/13 that was installed by installer
10:58:52AnLe2. i nnedto update to dev build as some devs ask in forum
10:58:58AnLeas i read here
10:59:09AnLethere is some sansa bootloader
10:59:28AnLei'am not quite sure, if i just copy бкщслищч
10:59:38wodzbootloaders != rockbox dev build
10:59:39AnLethen del it
10:59:56AnLethat unzip new dev build
11:00:00 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:00:08AnLeinto newly created rockbox
11:00:17wodzIf you are not explicitly ask to update bootloader don't do that
11:00:17AnLeso will it work?
11:00:27AnLei don't understand where is it
11:00:41AnLethat's why i ned to know what i can touch and what i don't
11:01:30AnLeSorry for the mistakes, a little worried.
11:01:36wodzto update the build you should 1) backup your current .rockbox folder (the one with 3.13 a + your config) 2) unzip on top of this new version from development build .zip
11:01:59wodzon unplug you should be asked if you want to reboot
11:02:14AnLebut, it's not looks like clear installation?
11:02:30AnLeare you shure that this is normal for testing dev build?
11:03:17AnLe;topicseen#new <- this theme cause me asking :)
11:04:02wodzyes I am sure. BTW. that was me who ask to test development build :-)
11:04:45AnLeyep, i see. also it's me who asked here when there is new 3.14 stable build )
11:04:57AnLeas i understand project stiill need testers
11:05:06wodzdesperately I would say
11:05:55AnLewell, i get new one sansa clip+, so now i can make some tests if it's needed on oldone.
11:06:03AnLeis there any roadmap for testers?
11:06:22AnLeforexample something cool was fixed and this thing must be checked
11:08:50wodzusb could be possibly fixed after squashing a few memory corruptions in theme engine
11:13:24AnLethat would be nice
11:14:30AnLehow can i easily rollback to stable release?
11:14:53AnLejust cope backup of my .rockbox?
11:17:28 Join onder`_ [0] (
11:17:51 Quit onder` (Ping timeout: 246 seconds)
11:17:56 Nick onder`_ is now known as onder` (
11:29:34wodzAnLe: yes
11:30:17AnLeoh, so simple.
11:30:30AnLethanks :)
11:31:27AnLeis there any sense test usb connections via some hubs or just directly to pc with win|lin would be enough?
11:32:44 Quit pixelma (Read error: Operation timed out)
11:32:48 Quit amiconn (Read error: Operation timed out)
11:33:23wodzPreviously it was known that some themes breaks usb connection reliably so this themes NOT breaking usb is what we want to get tested
11:35:17AnLeis it some known themes that have 100% hit this bug?
11:35:27AnLeor just test few themes that i like?
11:35:59wodzthe letter if you experienced usb - theme interference previously
11:37:48 Join Rower [0] (
11:39:34 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
11:39:43 Quit nosa-j (Ping timeout: 252 seconds)
11:40:04 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
11:46:01 Join nosa-j [0] (~m00k@
11:48:07 Quit amiconn (Disconnected by services)
11:48:07 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
11:48:10 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
11:48:10 Quit pixelma (Ping timeout: 260 seconds)
11:48:16 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
11:48:17 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
11:54:05AnLewodz: letter?
11:54:11AnLeliteral maybe?
11:55:18 Quit nosa-j (Ping timeout: 272 seconds)
11:56:54 Join nosa-j [0] (~m00k@
11:57:54 Quit amiconn (Disconnected by services)
11:57:54 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
11:57:56 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
11:58:12 Quit pixelma (Read error: Operation timed out)
11:58:56 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
12:00:21 Quit amiconn (Disconnected by services)
12:00:21 Quit pixelma (Disconnected by services)
12:00:21 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
12:00:21 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
12:00:21 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
12:00:24 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
12:00:52kugelgevaerts, JdGordon: alright, I got motivated to provide a better solution. I think have something workable
12:01:05gevaertsGreat :)
12:03:31 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
12:03:31 Quit pixelma (Disconnected by services)
12:03:32 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
12:03:51 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
12:03:51 Quit amiconn (Disconnected by services)
12:03:54 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
12:06:07 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
12:06:07 Quit amiconn (Disconnected by services)
12:06:08 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
12:06:08 Quit pixelma (Disconnected by services)
12:06:09 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
12:06:09 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
12:08:53 Quit amiconn (Disconnected by services)
12:08:53 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
12:11:04 Quit pixelma (Ping timeout: 252 seconds)
12:11:58 Join amiconn [0] (quassel@rockbox/developer/amiconn)
12:11:58 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
12:13:22 Quit amiconn_ (Ping timeout: 260 seconds)
12:13:46 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
12:13:46 Quit amiconn (Disconnected by services)
12:13:49 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
12:14:18 Quit pixelma (Read error: Operation timed out)
12:14:38 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
12:16:17 Quit pixelma (Read error: Operation timed out)
12:16:38 Quit amiconn (Read error: Operation timed out)
12:20:07 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
12:20:49 Join amiconn [0] (quassel@rockbox/developer/amiconn)
12:23:30JdGordonkugel: oh?
12:25:29 Quit pixelma (Disconnected by services)
12:25:29 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
12:25:30 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
12:26:32 Quit amiconn (Disconnected by services)
12:26:32 Join amiconn__ [0] (quassel@rockbox/developer/amiconn)
12:26:36 Nick amiconn__ is now known as amiconn (quassel@rockbox/developer/amiconn)
12:36:32kugelgevaerts, JdGordon: As briefly mentioned yesterday I think the best would be to hope the cache has everything needed (and provide a fallback for the very unlikely case where the assumption doesnt hold). I'm implementing just that
12:38:45 Quit amiconn (Disconnected by services)
12:38:45 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
12:38:49 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
12:38:49 Quit pixelma (Disconnected by services)
12:38:49 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
12:38:51 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
12:39:28PurlingNayukiAny way to revert fnt files to bdf?
12:45:34 Quit pixelma (Ping timeout: 276 seconds)
12:45:34 Quit amiconn (Ping timeout: 276 seconds)
12:47:05 Join mt [0] (
12:47:07 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
12:47:23 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
12:50:50 Join mt_ [0] (
12:52:52 Quit mt (Ping timeout: 252 seconds)
12:54:38***Saving seen data "./dancer.seen"
12:55:40kugelcopper: ping
13:07:20 Quit pixelma (Disconnected by services)
13:07:20 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
13:07:21 Quit amiconn (Disconnected by services)
13:07:21 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
13:07:22 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
13:09:08 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
13:09:10copperkugel: hai
13:09:19 Quit amiconn (Disconnected by services)
13:09:19 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
13:09:22 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
13:11:36 Quit amiconn (Disconnected by services)
13:11:36 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
13:11:38 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
13:12:06 Quit pixelma (Disconnected by services)
13:12:06 Join pixelma2 [0] (pixelma@rockbox/staff/pixelma)
13:12:07 Nick pixelma2 is now known as pixelma (pixelma@rockbox/staff/pixelma)
13:12:22 Quit pixelma_ (Ping timeout: 252 seconds)
13:13:39 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
13:13:40 Quit amiconn (Disconnected by services)
13:13:40 Quit pixelma (Disconnected by services)
13:13:40 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
13:13:42 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
13:13:42 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
13:14:52 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:14:52 Quit dfkt_ (Disconnected by services)
13:21:48 Quit nosa-j (Ping timeout: 272 seconds)
13:22:20 Quit maruk1 (Quit: Leaving.)
13:23:00 Join nosa-j [0] (~m00k@
13:24:28kugelcopper: does if (cache_only)
13:24:28kugel return NULL;
13:24:39kugeldoes FS #12892 still happen?
13:24:40fs-bluebot backdrop doesn't load when selecting a new theme that was just copied via USB (bugs, unconfirmed)
13:34:12 Quit nosa-j (Ping timeout: 272 seconds)
13:39:26kugelgevaerts, JdGordon: g#723 g#724
13:39:28fs-bluebotGerrit review #723 at : by Thomas Martitz (changes/23/723/1)
13:39:29fs-bluebotGerrit review #724 at : usb: Do not disable themes in the USB screen. by Thomas Martitz (changes/24/724/1)
13:40:36fs-bluebotBuild Server message: New build round started. Revision 80aac92, 249 builds, 32 clients.
13:41:09copperkugel: checking
13:42:43 Quit fs-bluebot (Remote host closed the connection)
13:43:02 Join fs-bluebot [0] (
13:43:31fs-bluebotBuild Server message: Build round completed after 175 seconds.
13:43:51copperkugel: can't reproduce it on either the Fuze+ or the iPod Classic
13:44:49copperI plugged in my iPod, copied the Xplorr theme, selected it in the theme list, loaded fine
13:45:15copperthen I tried to load my Clippy theme, but the Xplorr backdrop is still there
13:45:48copperthen I tried loading my Clippy theme a second time, and then the correct backdrop loaded
13:45:49gevaertskugel: I'll have a proper look tonight, but at first sight it looks ok
13:46:53gevaertsI'm wondering if we couldn't use a space character for the blank glyph though. It wouldn't be full width, but I don't know if that's needed, and it would save a (tiny, probably) amount of RAM. Not sure if it's worth it
13:47:49gevaertsI'm also not convinced that we won't need some sort of glyph preloading to force the cache to have more useful content before we close the file, but that's for later
13:47:56kugela single glyph is so tiny it doesn't matter
13:48:19gevaertsWell, on a 50 point font... :)
13:48:57kugelgevaerts: the second commit performs a UI update, which should preload all necessary glyphs. well, unless the sbs is extremely dynamic
13:49:12gevaertsAh, ok. That's probably enough then
13:50:19gevaertsAnd if not, we could still add things like preloading digits so clocks work properly. Let's leave that for when we know it's really needed though
13:50:29kugelassuming digits are competely in cache
13:51:44kugelon font load the iso8859-1 range is cached, including digits. it seems highly unlikely that digits get out of the cache
13:52:09gevaertsWell, themes can specify how big the cache should be
13:53:16gevaertsSo if they do specify e.g. 11 (enough for HH:MM:SS, i.e. all digits and the ':'), that will end up having the right content after ten seconds, but not from the iso8859-1 preload
13:53:24gevaertsWell, not necessarily
13:53:36kugelin theory, yes, although this setting is not a theme setting, and thus not exported by "save theme settings" and not accepted by the theme site
13:55:17kugelthe minimum is 50
13:58:33kugelbut preloading digits is just a matter of a dummy lcd_puts(0,0,"0123456789") call
14:01:19gevaertsIt's an option to %Fl
14:02:09kugeloh, awesome, a way for themes to screw the OS up
14:04:12kugelanyway, we can consider explicit preloading if it turns out required
14:05:27*gevaerts nods
14:05:36 Join pretty_function [0] (~sigBART@
14:05:48 Quit pretty_function (Remote host closed the connection)
14:24:23 Join onder`_ [0] (
14:24:24 Quit onder` (Ping timeout: 246 seconds)
14:24:44 Nick onder`_ is now known as onder` (
14:29:05chrisjjA git fail, in case anyone is interested:;a=tree;h=0a3a68414633eea492ddf9a2fe0de7b7058a3161;hb=0a3a68414633eea492ddf9a2fe0de7b7058a3161
14:29:31chrisjjWorkaround:;a=tree perhaps.
14:29:38toehser1I've seen weird "still have old backdrop, fixes itself on power cycle" also. Should "backdrop: -" in the cfg clear some other backdrop? It doesn't always.
14:30:16 Join Extricated [0] (516ca1ee@gateway/web/freenode/ip.
14:30:31ExtricatedHiya, if I install Rockbox on my iPod will I lose all of my music and have to re-add it?
14:34:42ExtricatedThanks for the help. /s
14:34:43 Quit Extricated (Client Quit)
14:36:12copperkugel: seems like FMS artwork sometimes crashes Rockbox
14:36:25copperthe sim segfaults, and the Fuze+ data aborts
14:36:47copperwhen switching FM presets
14:38:05copperI can still hear the radio, but the device is stuck on the data abort screen
14:39:52copperiPod Video sim too
14:41:28 Quit Zagor (Ping timeout: 252 seconds)
14:41:29toehser1PurlingNayuki: From what I saw the font conversion utilities are unidirectional to the fnt format. I would also like a way to go backwards from fnt (for me, mostly to be able to edit a single glyph or add one in toolsets that support other formats). An GUI editor that operates on fnt would do the trick of course but making a bidirectional converter would be much easier.
14:41:36 Join amayer [0] (
14:41:55 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
14:44:14toehser1(Stuff working when on USB sounds great, perhaps touchscreen devices can also be recognized as trackballs while you're at it?)
14:45:40kugelwow, just wow
14:46:45kugelI'm depressed, seeing what crap code I sometimes produce
14:48:46kugelcopper: fixed
14:48:47toehser1(Wait - I've got it - my clipzip could be a USB microphone / USB sound card! Why not? Can you read the mic/write the audio out/do USB? (jk))
14:48:49copperthat archive extracts to the root directory and contains a preset file and corresponding FM art (for most stations)
14:48:56fs-bluebotBuild Server message: New build round started. Revision 644d9ea, 249 builds, 32 clients.
14:48:57copperkugel: already?
14:49:39kugeltoehser1: pamaury had worked on that some time ago
14:49:56kugelcopper: thanks for that archive, it helped a lot
14:50:10toehser1I should have known someone here would also be tetched.
14:50:48copperkugel: did you forget to clear RAM usage or something?
14:50:59PurlingNayukitoehser1: Fnt format is actually not an original format by Rockbox?
14:51:34toehser1I think it is, but that doesn't affect the question, so I never gave that much thought.
14:51:53kugelcopper: ?
14:52:03toehser1What are you trying to do? Pretty much most of the fonts you can find in a source format to manipulate if you have to.
14:52:13copperanyway, fixed indeed, thanks :)
14:52:35chrisjjtoehser1: "could be a USB microphone / USB sound card!" Also nice would be remote UI access from PC to player.
14:53:07toehser1Or a remote debugger! The sim is nice, but remote debugging would be cool.
14:53:21 Join LinusN [0] (
14:53:58toehser1But if I understand my (minimal) skimming, nothing that needs flash access can happen when USB is active, so most of those things are off the table.
14:54:40***Saving seen data "./dancer.seen"
14:55:09kugelmost things, including the ones you mentioned, don't need filesystem access
14:55:35kugelfor cases where this is requried you could use a ramdisk
14:56:57toehser1There are docs around about how to use fontforge, and ways to find the original fonts that stuff came from. But I would still like an implementation of the use case "edit a rockbox font" from fnt=>something=>editor=>something=>fnt.
14:58:19toehser1kugel: the thread was saying that ram is often so limited that even a full font can't fit in memory, much less a full song, so I think things like remote UI access and probably even remote debugging are in practice off the table, if the disk can't be accessed while USB traffic is active.
14:59:07toehser1but the soundcard/mic idea or the trackball idea don't need disk of course
15:00:31kugelwell, some targets can't support everything
15:00:48kugelbut there are 64MB targets, which is plenty for toying around
15:01:44kugelalso, filesystem access is only blocked during UMS mode
15:02:06kugeli.e. no UMS -> full access
15:07:41PurlingNayukitoehser1: I want to get the source of font.
15:08:00PurlingNayukiIt isn't included in the source
15:08:26PurlingNayukiNor can be found by searching the file name.
15:10:19gevaertstoehser1: nothing that needs storage can run when USB Mass Storage is active. USB itself isn't the issue
15:10:31gevaertsIt's just that you can't access the same filesystem from both sides
15:13:38chrisjjtoehser1: Oh yes, remote debugger could bring this project on by miles!
15:14:05chrisjjThe sim is no substitute for the device code.
15:15:01chrisjjThis flash v. USB is a limitation only of the current RB code?
15:16:17gevaertsOf course not
15:16:17kugelchrisjj: re the playlist viewer option/manual thing
15:16:29toehser1You could access the same filesystem from both sides if one was in r/o mode with a flag to tell it to refresh when the other side wrote...
15:16:33chrisjjmmm... ?
15:16:49kugelchrisjj: at the time I made the comment the manual didn't reflect the code. later on I changed the code so that the manual is right again
15:16:49gevaertstoehser1: have fun implementing that in a reliable way on top of something like FAT
15:16:50toehser1PurlingNayuki: The source of WHICH font?
15:16:57chrisjjtoehser: great.
15:17:16chrisjjkugel: OK, I get it. Thanks.
15:17:31gevaertsThere are specialised filesystems that can do that, but none of the common ones do
15:18:23chrisjjkugel: kugel: BTW, your "Reason for closing: Fixed Additional comments about closing: fe73d75" refers to the wrong commit.
15:19:03kugelindeed, sorry about that
15:19:05toehser1It sort of depends - if one side is just reading mp3 and fnt and wps and cfg files that aren't changed by the other side, for example, there shouldn't be a problem, they aren't going to be moving around in the FAT. And, if the writes are infrequent enough from the writing side, the cost of just unmounting/remounting from the other side when told of an update may be acceptable.
15:19:16chrisjjkugel: Correct is;a=commit;h=2d4bf62032cf1d793abfbe73942ec050161844c1
15:19:47kugelright, I already changed the close msg
15:20:23gevaertstoehser1: you're free to run a defragmenter on the host
15:20:44gevaertsIt's really a huge can of worms
15:20:55gevaertsWhich is why if you want that sort of thing you do MTP
15:22:47chrisjjkugel: Thanks. BTW, odd that my "Reason for request" has appeared as a comment.
15:26:51PurlingNayukikugel: G#384 will be updated?
15:26:53fs-bluebotGerrit review #384 at : Touchscreen: Show a line separator in lists. by Thomas Martitz (changes/84/384/3)
15:27:28kugelPurlingNayuki: yes, later
15:27:55PurlingNayukiIt is by design to show on all targets or touchscreen targets only?
15:28:44kugelwas touchscreen-only at first but I changed my mind
15:29:15toehser1Not sure I agree about the can of worms aspect. But I'm also not sure that the use cases are valuable enough to make it worth doing. I'm more interested in the idea of remote debugging. But another way to do it, instead of giving the host direct access, give the host access through the os of the player or reroute the player's access through the host OS - if everything goes through one side, the can of worms would be a lot smaller. But I will say, if
15:29:54gevaertsYou don't need disk access for remote debugging
15:30:26toehser1Right - I was mixing two conversations - caffeine registers are still low here in -4 TZ.
15:30:49toehser1I mean -5
15:32:22PurlingNayukikugel: Then I'll just wait :)
15:32:47PurlingNayukiWould try to implement lcd_get_dpi for all targets if you said no
15:33:24kugelPurlingNayuki: I only need the lcd sizes of all of our targets (in inch)
15:34:06kugeli build a script to calculate the dpi and add the needed line to config/$target.h
15:34:36toehser1PurlingNayuki: Which font are you trying to find source for?
15:34:50PurlingNayukiThat's espy asian
15:35:12kugelPurlingNayuki: if want to help you could collect some for me
15:36:04gevaertskugel: make sure lebellium knows you need this :)
15:36:18PurlingNayukiCollect what? LCD size?
15:36:42kugel"I only need the lcd sizes of all of our targets (in inch)"
15:37:41PurlingNayukiThat sounds not difficult.
15:37:55kugelright, it only takes a lot of time
15:38:10PurlingNayukiJust push a sample to gerrit and I'll help
15:38:49PurlingNayukiFor I'm actually interested and using it. Also added this to my published rockbox version.
15:38:58kugeli just need a list "$target: XX inch"
15:39:10kugelI'll feed my script with that (manually)
15:41:18PurlingNayukiJust, nasty and dumplicated work.
15:42:48PurlingNayukitoehser1: Any results for espy asian? I once thought it was zpix but it wasnt.
15:45:57kugelPurlingNayuki: could also make the list machine readable; anyway I'm going to need such a list
15:46:56PurlingNayukiShouldn't we add this to $target.h?
15:50:31chrisjjkugel: On "at the time I made the comment the manual didn't reflect the code." understood, but I hope we agree, that comment "however the manual is still wrong indeed." was inaccurate. The manual was right - it was the code that was wrong.
15:53:17toehser1The espy fonts I think came from Apple Newton originally, then at some point were made legal to use, and used for daps.
15:56:27PurlingNayukiHmmm let me take a look
15:56:57PurlingNayukiBut it's with unicode support
15:57:26PurlingNayukiThought to be full unicode support, including BIG5, GBK and SJIS
15:57:56PurlingNayukiIs it espy sans?
16:04:34pamaurytoehser1: yeah I worked on making rockbox a usb soundcard and even making the Fuze+ touchpad a HID touchpad, but both work were never finished and are not probably out of sync, even though they kind of worked
16:05:38wodzpamaury: you should really finish usb audio work. This could be killer feature of rockboxed dap
16:06:21pamauryyeah agree
16:06:41PurlingNayukiAgree too
16:06:50toehser1I think there was an "apple guide localization kit" for example. Espy sans probably doesn't have what you want, probably is a subset. I see references to it being in eworld fonts also. I think what you're looking for is versions of apple stuff that were localized, way back when.
16:08:04pamaurybut first I said I would finish my ongoing work on some DAPs
16:08:16pamaurylike make plugins works, update manuals and so on
16:08:24toehser1Clip has a pretty good microphone to my ears, and definitely good DA, if we could introduce a better usb mic and da output at a lower price than dedicated solutions, it would probably make a few sit up and notice.
16:09:23toehser1Disclaimer: I hate apple and have never owned an apple product. So I'm probably a sub-optimal person to fish this one.
16:12:20toehser1But I see references to things like "apple guide maker" and "apple guide localization" and "eworld fonts are espy fonts" from way back when, making me suspect that this is where to go... (Any chance some other font could meet your needs? What about espy is special?)
16:12:45PurlingNayukitoehser1: I think the espy asian font is a mixed-up font. Look at its name, espy + asian :)
16:12:56PurlingNayukiShould really try to find out what asian font is.
16:13:14wodzpamaury: you should really leave the easy work to others.
16:15:03chrisjjpamaury: USB audio the other way too would be nice. Device −−> PC. Good for a future remote access mode. Combined with remote debugging, this would be a killer device IDE.
16:17:52toehser1I think someone just ran the localization on an asian machine using the apple guide (help system) development environment. All you need is a 1995 Mac development environment from the country you want...
16:20:00toehser1Another approach would be to try to track down the theme author who first included it, from theme comments and such, if you can dig up an email address, ask them...
16:20:03PurlingNayukiI don't think there would an apple designed just for Hongkong :(
16:20:56toehser1I think the OS was in fact localized, and that affected the generation of apple guide help files and their associated font file, from what I see.
16:21:37toehser1but, again- why that font? Are there similar open source fonts that are easier?
16:22:10PurlingNayukiIMO that's the best font
16:22:20PurlingNayukiEspecially for that size
16:22:50PurlingNayukiIt seems to be designed just for small screen targets running rockbox
16:26:21kugelPurlingNayuki: that's the plan, as I said
16:26:55kugelif the list would be "<target.h>: XX inch" I could automate the whole process
16:29:42toehser1I like the noto fonts. What are you trying to do with it- add glyphs or something?
16:30:21toehser1When you say it is the "best font", you mean "out of the ones you happen to have in your player", or, "in the world"?
16:30:57pamaurywodz: no one does it otherwise :-(
16:31:14pamauryAnd I don't want to leave ports in a half finished state
16:31:47pamaurychrisjj: what do you mean the other way ? USB audio is full duplex: you can play and record
16:32:01PurlingNayukitoehser1: among all the fonts that have (full) unicode support and under 12px
16:33:41chrisjjpamaury: What I mean is that the audio from the device sounds through the PC.
16:33:42toehser1When you say "all the fonts", do you mean "all the fonts on your player" or "all the fonts in the world"?
16:34:02pamaurychrisjj: that's recording no ?
16:34:18toehser1We really want 3 audios - player (playing) to pc, player (mic) to pc, and pc to player.
16:34:18 Quit Strife89 (Read error: Operation timed out)
16:34:39PurlingNayukiOK I surrender
16:34:41chrisjjThere's a fourth, but let's not be greedy :)
16:35:36 Quit mt_ (Ping timeout: 272 seconds)
16:36:27pamaurychrisjj: I don't get it ? What do you expect USB audio to be used for if it's neither playback or recording ?
16:37:52toehser1I just mean to say, there are probably fonts with full unicode support and under 12px that you haven't looked at, that are MUCH easier to find/work with that espy-asian... I would spend time looking from that angle, before the espy-asian one... have you looked at the ones Greg Till converted, at ?
16:38:13chrisjjpamary: It is playback. You said "I worked on making rockbox a usb soundcard" - that's tracks on PC sounding on device, I assume. My suggestion is the other way (direction): tracks on device sounding on PC.
16:39:34pamauryErr, I still don't understand, anyway I'm not working on it right now
16:39:41toehser1what is the fourth? Yes, pc apps see the player as an output device, yes, pc apps see the player as an input (from it's playing of files), yes, pc apps see the player as an input (access to it's microphone)... that is 3
16:39:52PurlingNayukitoehser1: Actually, yes and no.
16:40:19PurlingNayukiDownloaded before but they're all anti-aliased right?
16:40:55toehser1Those files are from known sources, you can go back to truetype or bdf, edit, reconvert, whatever. We have all the pieces.
16:40:58gevaertschrisjj: as far as USB audio is concerned, that's exactly the same as recordingf
16:41:17PurlingNayukiAnd most of them don't have chinese support
16:42:03toehser1I think converting from AA to normal isn't hard in something like fontforge, is it? It is loss of information. NoTo is supposed to have EVERYTHING support - the "noto" means "no tofu" from a reference to the "squares" you see when a font doesn't have a character - so I think it is probably pretty complete as far as languages.
16:42:10toehser1Why don't you want AA?
16:43:21PurlingNayukiBecause of the small screen on sansa zip
16:44:06toehser1Huh? AA is better on zip. I have zip, an looked at both side by side, before choosing fonts for my themes. I'll take NoTo on the Zip over ANY non-AA font. Looks better.
16:44:06PurlingNayukiUse AA fonts on GBS, Fuze and iPod Nano2 but just don't like the unclearness on clip zip
16:44:33PurlingNayukiBetter, yes, but not for chinese.
16:45:00toehser1I see - the thin complex lines aren't done right in the AA. AA could have left those lines without fuzz, but didn't.
16:45:09gevaertsAA is a matter of opinion. It's not *objectively* better or worse in any case
16:45:38toehser1Really just bad authoring - not AA vs. not AA, just that the person who designed the chinese glyphs made bad choices.
16:46:24toehser1Well - here is what I would do. Look at the BDF=>FNT source. If it looks straightforward to add reverse-conversion, do that. If not, bribe the author to do it.
16:47:06toehser1AA is better in that the AA font can have hard edges (exactly reproduce non-AA), but not the other way.
16:47:07PurlingNayukiPlus hongkong uses traditional Chinese, which makes it more difficult to write and display while it looks more beautiful
16:47:38toehser1AA fonts can display a superset of what non-AA fonts can display.
16:47:55PurlingNayukiHow do you share picture in IRC? Show you a picture and you'll find that non-AA is also not bad.
16:47:55toehser1Any given AA font, though, I agree, isn't inherently better than any given non-AA font.
16:48:16PurlingNayukiAgain, in terms of Chinese displaying.
16:48:36toehser1People use stuff like pastebin for text, must be an equivalent for images. Doesn't matter though - I get the idea.
16:50:04PurlingNayukiIf I use English only, I'll also definitely choose AA fonts −− but actually I'm not.
16:54:41***Saving seen data "./dancer.seen"
16:58:07 Quit Zagor (Quit: Clint excited)
16:59:32chrisjjtoehser1: "player (playing) to pc, player (mic) to pc, and pc to player." pc mic to player.
17:00:34 Quit fragilematter (Quit: Leaving.)
17:00:50chrisjjPerhaps best to consider this as player (whether files, mic or any other) to PC, and PC (whether files, mic or any other) to player.
17:01:10gevaertschrisjj: that's what everyone but you is doing, yes
17:01:38chrisjj"People use stuff like pastebin for text, must be an equivalent for images." Perhaps try .
17:06:27 Join krabador [0] (~krabador@unaffiliated/krabador)
17:17:10 Quit foolsh (Ping timeout: 265 seconds)
17:26:40toehser1chrissjj: That is a routing thing that the player doesn't need to see - if I have an output to the player, I can just use jack routing to route any pc source to that channel. Why would rockbox need to see 2 stereo channels to the pc? But, I _could_ see a case for 2 channels to the pc from the player: Say: karaoke track on the mp3 file, and you're singing with it using the player's microphone, and want both those channels going to the pc...
17:27:16toehser1(By "to" I meant "from" at one point there)
17:28:01kugeltoehser1: there's imagebin
17:29:01wodztoehser1: if you think about playing something on the dap, using dap's microphone and sending mix of this to the pc through USB you are way too futuristic :-)
17:29:59toehser1Oh - I wasn't thinking about sending a mix - I was thinking about sending both streams, for possible mixing on the PC.
17:30:53wodztoehser1: not possible either
17:30:58toehser1The only use case I think I would personally care about would be the USB microphone case.
17:31:02wodzat least not with current daps
17:31:17toehser1Why not? CPU limits, USB chip limits?
17:31:38wodzafaik you cannot record and play something at the same time
17:31:45toehser1Or the same audio chip path for the mic and the audio?
17:31:53wodzdigital part, yes
17:32:26wodzat least for vast majority of daps
17:33:10toehser1Most PCs already have D-A outputs on speakers or headphone jacks, anyway, so the PC=USB=>Player=>audio-out case isn't very interesting, to me.
17:34:25kugelrockbox doesnt allow simultaneous playback and recording
17:36:49kugelpamaury: re FS #10916: I found the vhci driver is now in mainline (though staging), could revive that task?
17:36:50fs-bluebot Add usb capability to simulator on Linux hosts (patches, new)
17:37:14pamauryoh really ? that's great news !
17:37:37pamaurydamn, that was already 4 years ago :-/
17:37:46kugelit's under USB/IP, somewhat hard to find
17:37:53kugel(in menuconfig)
17:38:25pamauryI'm not so sure the usbip stuff will ever make it to mainline
17:43:18pamauryah yeah you are right, I wonder how does it compare with the usb vhci stuff on sourceforge
17:45:15pamauryapparently it is not the same
17:45:27 Quit wodz (Quit: Leaving)
17:46:59pamaurybut yeah this thing was very useful to stress the simulator with usb
17:49:17pamaurykugel: if you are interested, I can see if I can revive it. A priori it should not have changed much
18:08:09chrisjjtoehser: no I wasn't thinking that futuristic! :)
18:11:00saratogadoes AMSv2 have a bootloader USB mode? i've seen a few people mention that USB is more stable if the player is booted first
18:11:06chrisjjtoehser: My interest is only in in player audio going to PC speakers, since that's what's needed for RDP-style remote access, that being very useful for development.
18:13:03 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:14:18 Join Baranko [0] (
18:15:58Barankoguys, i take devbuild, but there is still usb problems:,43526.0.html
18:16:24Barankoi have it when it hub connected and when i transfer files.
18:16:35Barankois there anything like logs that can help you?
18:21:55 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:27:14 Join saratoga_ [0] (ac3817b8@gateway/web/freenode/ip.
18:32:11 Join saratoga__ [0] (ac3817b8@gateway/web/freenode/ip.
18:32:13 Quit saratoga_ (Ping timeout: 272 seconds)
18:32:37saratoga__the log to disk code could be pretty useful for USB debugging
18:33:08saratoga__it should be able to store debug info and dump to disk after disconnect
18:36:29 Quit saratoga__ (Client Quit)
18:41:17Barankoit can be enable somewhere in sansa or what? )
18:43:55 Join rela [0] (~x@pdpc/supporter/active/rela)
18:45:53 Join Airwave [0] (~Airwave@
18:51:55kugelpamaury: I'm sure interested, especially considering the number of theme/usb related bugs in the last months
18:54:44***Saving seen data "./dancer.seen"
18:57:01 Part LinusN
18:59:57*kugel squashed another bug
19:03:22 Join Strife89 [0] (~Strife89@2602:306:250d:18b9:e42e:ed47:3ab3:c4bf)
19:13:36copperwhat bug?
19:14:51toehser1I have a couple of clip+ where either the power button broke or something else such that they aren't useful as players really, but I could play with JTAG on them - any recommendations on USB-JTAG hardware?
19:28:30 Quit ParkerR (Ping timeout: 240 seconds)
19:29:06 Join ParkerR [0] (ParkerR@unaffiliated/parkerr)
19:29:49 Quit habys (Ping timeout: 240 seconds)
19:30:01 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
19:30:13pamaurykugel: ok, I'll try to have a quick look tomorrow
19:37:48toehser1Hmmm... I see that a $5 arduino clone can be a USB-JTAG adapter... I think I'll investigate that before going the 80-Euro route...
19:39:00 Quit tchan (Quit: WeeChat 0.4.2)
19:40:16 Quit rela (Read error: Connection reset by peer)
19:42:09 Quit pamaury (Ping timeout: 272 seconds)
19:45:48 Quit Baranko (Quit: Ухожу я от вас)
19:52:16 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
20:03:00 Quit kugel (Ping timeout: 245 seconds)
20:13:02 Quit krabador (Ping timeout: 260 seconds)
20:15:51 Join krabador [0] (~krabador@unaffiliated/krabador)
20:31:27 Join einhirn [0] (
20:36:04 Quit einhirn (Ping timeout: 248 seconds)
20:43:49 Join einhirn [0] (
20:49:12 Quit fyre^OS (Ping timeout: 272 seconds)
20:54:46***Saving seen data "./dancer.seen"
21:13:45 Join mt_ [0] (
21:14:34 Join wodz [0] (
21:15:27 Quit Rower (Quit: Hmmm...)
21:17:01wodzkugel: (log) what about reds?
21:28:55 Join habys [0] (~luke@
21:34:25 Quit mt_ (Read error: Connection reset by peer)
21:38:48 Join krabador_ [0] (
21:40:46 Quit krabador (Ping timeout: 260 seconds)
21:48:18 Quit krabador_ (Remote host closed the connection)
22:06:55 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
22:06:55 Quit pixelma (Disconnected by services)
22:06:56 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
22:06:57 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
22:06:57 Quit amiconn (Disconnected by services)
22:07:00 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
22:15:15 Join Zarggg [0] (
22:30:20 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
22:32:31 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
22:33:06 Quit fs-bluebot (Ping timeout: 260 seconds)
22:33:34 Quit bluebrother^ (Ping timeout: 260 seconds)
22:34:31 Join fs-bluebot [0] (
22:35:27 Join kugel [0] (
22:35:27 Quit kugel (Changing host)
22:35:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:44:33kugelgevaerts: did you already have a look at the font thingy?
22:46:37 Quit MarcAndersen (Quit: I was using NightOwl 0.2.)
22:51:18wodzkugel: still some reds
22:54:48***Saving seen data "./dancer.seen"
23:00:45 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:10:27 Quit GodEater (Ping timeout: 246 seconds)
23:14:09 Join GodEater [0] (
23:14:09 Quit GodEater (Changing host)
23:14:09 Join GodEater [0] (~whoknows@rockbox/staff/GodEater)
23:16:58 Quit wodz (Quit: Leaving)
23:18:32 Quit amayer (Quit: Leaving)
23:30:17gevaertskugel: I can't find anything I dislike :)
23:30:28gevaertsIt all looks reasonable to me
23:30:51 Join MarcAndersen [0] (
23:31:04gevaertsOf course, take that with the grain of salt that matches my knowledge of apps/ code in general
23:32:48kugelgevaerts: most changes are in firmware :)
23:33:18kugelif you looked at it you could press the review button on gerrit :p
23:33:58*gevaerts will
23:36:29kugelgevaerts: what target did you test on?
23:36:34gevaertsipod video
23:36:49gevaertsI didn't do very intensive tests though
23:38:56gevaertsSo now I'll have to work on having the usb code maintain statistics about transfer speed and volume, so we can have a theme tag to show those
23:39:15kugelright :)
23:39:31gevaertsI mean, apart from that, what possible use could a skinnable USB screen be? ;)
23:40:35kugelit's not skinnable yet, it just doesnt disable the sbs anymore
23:41:03gevaertsWell, that means you can probably do creative sbs stuff and pretend it's skinnable
23:41:34kugelthat always works yes :)
23:41:51kugelalthough I think %cs doesnt indicate the usb screen yet, but that's easy
23:42:54 Quit bertrik (Remote host closed the connection)
23:43:40gevaertsYes, you do need that
23:44:24kugelworks on fuzev2 too with an sbs-enabled theme
23:45:09kugelI also tested the fallback to blank glyph. I'll just push it
23:46:57 Join onder`_ [0] (
23:47:35 Quit onder` (Ping timeout: 245 seconds)
23:47:41 Nick onder`_ is now known as onder` (
23:50:07kugeldo we have a manual section about usb
23:50:43gevaertsNot a specific section, no
23:51:54kugelthen there is no place to put a note about the possiblity of missing characters in the sbs during usb
23:54:08gevaertsI'd consider adding such a note to the theming secions
23:55:33[Saint]Im not even sure why there need be missing/replaced chars.
23:56:06gevaertsBecause of variable text
23:56:14[Saint]Isn't it a lot easier to just check if the required fonts fit in RAM and fall back to sysfont completely?
23:56:27[Saint]rather than doing it piece meal.
23:56:36[Saint]at least it'll look consistent.
23:56:37gevaertsI don't think so, no
23:56:53gevaertsThe way this is implemented is fairly straightforward
23:57:19kugel[Saint]: sysfont is shit
23:57:40[Saint]so is never knowing what glyphs youll have.
23:57:43gevaertsAnd I think that while it may drop a few glyph right now, we can probably make that very improbable reasonably easily if needed
23:57:48[Saint]thats terrible.
23:58:07gevaertsWe'll just blame the font's unicode coverage :)
23:58:15[Saint]plenty of targets have no need for this.
23:58:32kugelin practice everything shown in the USB screen will be in cache
23:58:38[Saint]we /could/ stuff all fonts in RAM on a lot of targets.
23:58:47[Saint]do this trickery with lowmem.

Previous day | Next day