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 2015-01-09

00:00:25[Franklin]has anyone thought about integrating PLA with the plugin API?
00:03:11 Quit bertrik (Remote host closed the connection)
00:31:23alexbobpwhere can I find more information about when the ipod control port stuff will work?
00:31:34alexbobp[Saint] said something cryptic about how it's a miracle it ever works
00:32:15[Saint]You could've asked me. ;)
00:32:41[Saint]This wiki page details the accessories that are known to be functional, either in full, or in part.
00:33:00[Saint]Please add to it if you find some accessories not listed therein.
00:35:26***Saving seen data "./dancer.seen"
00:36:09alexbobp[Saint]: I missed your message at the time and highlighted you later but I think you didn't notice
00:36:12alexbobpthanks for the link :)
00:37:12 Quit wodz (Quit: Leaving)
00:38:26alexbobp"and is only implemented for the 4G, 5G and Nano 1G"
00:38:31alexbobpthat explains it :( I have mini 2g
00:39:27alexbobp[Saint]: I hope this isn't getting into obnoxious question territory but any idea why it's not implemented for mini?
00:40:03[Saint]I don't think we know how to use the serial port on the mini
00:40:21alexbobpaw I see
00:40:46[Saint]with the amount of reverse engineering of a proprietary protocol this took, its really is amazing that there'
00:40:53[Saint]s as much support as there is.
00:41:03alexbobpyeah I agree. I also curse apple for their proprietaryness...
00:42:33[Saint]There's a metric shittone of awesome things IAP can do, and several different protocol versions, but Apple isn't in the business of giving away the keys to the kingdom, as it were.
00:42:48[Saint]Unless you pay to license it and sign an NDA.
00:44:09 Quit xorly (Ping timeout: 252 seconds)
00:45:02alexbobpwhich you would be violating the moment you implement the features in open source code, I assume
00:45:08[Saint]One cool thing I've seen before is accessories being able to grab the iPod screen and display arbitrary images. On one of my old car stereo head units, with the original iPod firmware on an iPod Video, it would take over the iPod screen entirely and display a Kenwood logo.
00:45:23alexbobpit's just sad that Apple considers control over one's own purchased hardware, keys to the kingdom
00:46:04alexbobpalso taking over the ipod screen to display their own logo sounds douchey and annoying, but I get the idea for how that feature could be useful :P
00:46:36[Saint]I can understand why they want to control the experience, but it is frustrating from our point of view.
00:46:56[Saint]They mainly don't want generic accessories providing a shitty sub-par user experience.
00:47:12[Saint]Its understandable, but frustrating for us.
00:47:23alexbobpI don't understand how their current system prevents that
00:47:45alexbobpis the assumption that if mfgs have to jump through hoops to license the protocol, they will tkae it more seriously?
00:50:03 Join xorly [0] (
01:01:59[Saint]Apple gets to step in and look over their shoulders to ensure a quality experience is provided
01:12:20 Quit RiD (Read error: Connection reset by peer)
01:20:52alexbobp[Saint]: apple should come look over rockbox's shoulder then ;)
01:21:12[Saint]I would be very surprised if they haven't.
01:21:15alexbobpor I dunno, provide basic no-brainer functionality in their own firmware, like I dunno, playing files that are on the device. or editing playlists.
01:21:18[Saint]Incredibly so.
01:21:32alexbobpah, you think that's why later ipods are so hard to port to?
01:22:21[Saint]Not specifically, but I would be very, very, very surprised if they didn't know what we were up to.
01:22:37alexbobpoh I'm sure they are aware
01:23:04alexbobpwhen I say looking over your shoulders... I mean talking to devs about what they think would be a good rockbox user experience and what they want changed :P
01:23:57alexbobpI trust rockbox devs a hell of a lot more than apple to give me a quality experience, but they *do* have good ui experts and more eyes can't hurt
01:28:22[Saint]Regarding their device security, I would guess their heightened efforts if the later devices came more as a result of the iPhone jailbreaking community rather than us.
01:28:47[Saint]The later iPods are much much much closer to an iPhone than they are an iPod.
01:29:16alexbobpoh well... I guess that's probably true
01:29:30alexbobpthough that doesn't excuse them. I agree with iphone jailbreaking just as much as I agree with rockboxing :P
01:32:11 Quit chrisb (Ping timeout: 264 seconds)
01:42:25 Join chrisb [0] (
01:47:30foolsh[Franklin]: Yeah, the old console racing games usually had one resolution to deal with, so they could get away with scanline driven drawing engines
01:48:33 Quit thomasjfox (Quit: Konversation terminated!)
01:52:57[Franklin]foolsh: ... and now the more I think, the more I want to use a scanline-based engine :)
01:53:57foolshyou'd only have to work out the math on the fly, but it's do-able no doubt
01:54:27[Franklin]though making a full-blown 3d engine for rockbox would be kinda fun...
02:01:26 Quit xorly (Ping timeout: 264 seconds)
02:01:55foolsh[Franklin]: I had few more thoughts on it last night, the Y position (on screen) of an odject indicates it's Z value. Z and Y are linked in a non-liniear way (simply with a ^2 I believe is a close approximation), with farther objects appearing to approuch slower than near objects, that seem to move more quickly.
02:02:03foolshAs an object appears at the horizan (generally screen_height/2) it's Z value should be much less than 1, and closer to 0. This way you can translate the Object_width and Object_height by simply dividing them each by Z.
02:02:09foolshAs the object approuches the bottem of the screen (closer to the viewer) it's Z value shold approunch a value of 1 or greater, this way again the Object_width and Object_height can be simply translates by the divition with Z.
02:02:42[Franklin]what about the x-coordinate?
02:02:44[Franklin]is that linear?
02:03:38foolshyes unless you tilt the view angle it stays the same
02:04:03[Franklin]that doesn't make sense
02:04:14[Franklin]when you drive by something by the side of the road, it gets faster as you get closer
02:04:25[Franklin]faster horizontally
02:04:27foolshwell the width of the road is also determined by Z
02:04:53foolshmaybe I misunderstood
02:05:55*[Franklin] is by no means a graphics expert! :D
02:05:55 Quit AlexP (Read error: Connection reset by peer)
02:11:22[Franklin]I guess I'll just do exactly like says to do
02:12:47[Franklin]aha read this: " Though we now know the distance of each row of the screen, it may also be useful to cache either the width of the road or scale factor for each line. The scaling factor would just be the inverse of the distance, adjusted so that the value is 1 on the line which the player's car graphic spends the most time. This can then be used to scale sprites which are on a given line, or to find wh
02:12:54[Franklin]at the width of the road is."
02:12:59[Franklin]just what I need
02:13:18[Franklin]this implies a scanline method, so scanlines it is!
02:15:03foolshThat's what I was talking about :) divide it by Z where Z<1 when it far and Z>1 as it gets nearer, yes when you approuch something close, it appears to move more quickly as you pass it, thats what I said, I had to reread it :) to make sure, but the distance side-to-side that an object can move is reduced on screen by dividing by Z the father it is from the viewer, it would have nothing to do how fast it can
02:15:05foolshmove from one side of the road to other, that should be constant. Of course learning is always best when it's a journey of self discovery :)
02:15:57[Franklin]I can be an idiot sometimes...
02:19:30[Saint]No comment.
02:23:01 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
02:26:26[Franklin]well... it doesn't look /that/ bad :D
02:26:38 Quit advcomp2019__ (Ping timeout: 264 seconds)
02:26:48[Franklin]just a blocky road that rapidly recedes into the distance
02:27:11 Quit krabador (Quit: Take the time.)
02:28:08 Join krabador [0] (~krabador_@unaffiliated/krabador)
02:29:32[Franklin]foolsh: take a look at the latest patch set
02:29:58[Franklin]certainly not the best looking code... and not the best working code :)
02:30:21[Franklin]but things are starting to take shape now
02:35:29***Saving seen data "./dancer.seen"
02:49:55 Quit ZincAlloy (Quit: Leaving.)
02:53:18foolsh[Franklin]: Yes it is
02:55:32*[Franklin] really finds it hard to believe that there's not even 1 good example in C of scanline rendering
02:56:03foolshmust be some libsvga code some where out there that does this
02:56:14[Franklin]yea, that's my guess
03:02:49foolshyou can render it in "blocks" of scan lines, it would not look as smooth as a native resolution
03:04:14foolshHow can I put this... hang on
03:06:05foolshI guess, what I mean is choose an arbitrary number of horizonal scan lines you want the engine to have (lowest resolution target) and scale them up in size (vertical thickness) for higher resolution targets
03:06:57[Franklin]I'm thinking of the number of scanlines being dependent on screen size
03:07:09foolshThen I would do it with a vector engine
03:07:20foolshpoints and fills
03:07:39*[Franklin] is sliding in that direction anyways
03:07:44foolshor work out the math for it the other way
03:07:51[Franklin]pure scanlines don't let you do much
03:09:05[Franklin] seems to have a nice JS engine
03:25:14foolshthose sprites are butt ugly ;)
03:25:36 Quit krabador (Quit: Take the time.)
03:26:00[Franklin]better than what I have :P
03:36:14 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
03:36:43 Quit chrisb (Remote host closed the connection)
03:37:24 Quit [Franklin] (Quit: Lost terminal)
03:37:33 Join charlie [0] (~c@unaffiliated/charlie)
03:39:06 Quit charlie (Read error: Connection reset by peer)
03:41:20 Join charlie [0] (~c@unaffiliated/charlie)
04:10:09 Join chrisb [0] (
04:26:21 Quit chrisb (Ping timeout: 244 seconds)
04:35:33***Saving seen data "./dancer.seen"
04:47:10 Nick michaelni_ is now known as michaelni (
04:58:37 Quit the-kyle (Remote host closed the connection)
05:06:11 Quit bp0 (Quit: Leaving)
05:08:39 Join chrisb [0] (
05:12:51 Quit JanC (Ping timeout: 264 seconds)
05:13:14 Quit chrisb (Ping timeout: 264 seconds)
05:13:41 Quit TheSeven (Ping timeout: 244 seconds)
05:15:01 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:26:00 Join JanC [0] (~janc@lugwv/member/JanC)
05:40:26 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
06:11:59 Nick cmhobbs is now known as nilmethod (~cmhobbs@fsf/member/cmhobbs)
06:13:01 Nick nilmethod is now known as cmhobbs (~cmhobbs@fsf/member/cmhobbs)
06:13:53 Quit cmhobbs (Quit: Leaving)
06:14:06 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
06:35:37***Saving seen data "./dancer.seen"
06:57:02 Quit cmhobbs (Remote host closed the connection)
07:01:14 Quit dfkt (Read error: Connection reset by peer)
07:01:43 Join dfkt [0] (dfkt@unaffiliated/dfkt)
07:36:29 Quit Provel (Ping timeout: 244 seconds)
08:00:54 Join LinusN [0] (
08:25:19 Join ender` [0] (
08:35:39***Saving seen data "./dancer.seen"
09:03:03 Join xorly [0] (
09:08:23 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:13:06 Join einhirn [0] (
09:13:46 Join Provel [0] (
09:16:55 Join petur [0] (~petur@rockbox/developer/petur)
09:44:23 Quit pamaury (Ping timeout: 245 seconds)
09:54:24 Quit xorly (Ping timeout: 264 seconds)
09:57:21 Quit Jinx (Ping timeout: 245 seconds)
10:01:53 Quit einhirn (Ping timeout: 245 seconds)
10:03:51 Quit Ivoah_ (Ping timeout: 264 seconds)
10:16:35 Join pamaury [0] (
10:16:35 Quit pamaury (Changing host)
10:16:35 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:35:43***Saving seen data "./dancer.seen"
10:40:50 Quit Provel (Quit: Provel)
11:00:18gevaertsscorche|sh: I'd like to install , but I need your help with that (the installation obviously needs to write files, which you have sensibly disabled)
11:01:22gevaertsIf we have that, we can count posts in all forums, then update user post counts, then get rid of all old users with *really* 0 posts and no recent login
11:03:02gevaertsThe reason I'm asking is that we seem to have some spam with from old accounts, some of which apparently have positive post counts but no actual posts
11:03:13gevaertsSo cleaning that up would help
11:04:52 Join AlexP [0] (~alex@rockbox/staff/AlexP)
11:05:46gevaertsAfter that, I'm thinking of cleaning out the removed posts
11:10:38 Join nialv7 [0] (~nialv7@
11:44:16 Quit cronix (Ping timeout: 272 seconds)
12:01:43 Join cronix [0] (
12:23:59sobukusscreen -r
12:24:15sobukusSorry. Network lagging badly.
12:35:46***Saving seen data "./dancer.seen"
13:06:51 Quit AlexP (Remote host closed the connection)
13:42:28 Join AlexP [0] (~alex@rockbox/staff/AlexP)
14:00:31 Join Ivoah [0] (
14:15:22 Join einhirn [0] (
14:23:59 Join Guest66888 [0] (Slayer@
14:26:42 Quit Guest75680 (Ping timeout: 244 seconds)
14:29:26 Join Guest75680 [0] (Slayer@
14:32:23 Quit Guest66888 (Ping timeout: 244 seconds)
14:34:37 Join webmeister [0] (webmeister@unaffiliated/webmeister)
14:35:49***Saving seen data "./dancer.seen"
14:36:31 Quit kvieta (Ping timeout: 245 seconds)
14:39:51 Join kvieta [0] (~kvieta@
14:39:51 Quit kvieta (Excess Flood)
14:40:21 Join kvieta [0] (~kvieta@
14:40:21 Quit kvieta (Excess Flood)
14:41:51 Join kvieta [0] (~kvieta@
14:41:51 Quit kvieta (Excess Flood)
14:42:21 Join kvieta [0] (~kvieta@
14:42:21 Quit kvieta (Excess Flood)
14:42:25 Join krabador [0] (~krabador_@unaffiliated/krabador)
14:42:41 Join Rower [0] (
14:42:51 Join kvieta [0] (~kvieta@
14:42:51 Quit kvieta (Excess Flood)
14:43:33gevaertsscorche|sh: actually, it looks like either someone stole a bunch of 2 year old credentials, or someone was rather far-sighted two years ago and set up a lot of sleeper accounts, or there's a bug in smf they're coming through
14:43:51gevaertsAll of these have non-zero post counts without having actual posts...
14:44:21 Join kvieta [0] (~kvieta@
14:45:14gevaertsI wouldn't mind seeing our smf setup patched properly
14:48:16gevaertsIt's of course possible that those are accounts made back in one of the earlier spam wave, and that my patches back then didn't reset post count properly
14:48:23 Join wodz [0] (
14:49:07wodzgevaerts: Could you look at g#1130?
14:49:11fs-bluebotGerrit review #1130 at : Get rid of stupid _backlight_* function names by Marcin Bukat
14:55:58gevaertswodz: I don't have much of an opinion on that, I think
14:57:22wodzgevaerts: You don't have as in 'I don't care at all' or as in 'I don't have time to look at'?
15:05:26 Part LinusN
15:09:54 Quit Rower (Quit: Hmmm...)
15:10:18 Join edhelas [0] (~edhelas@
15:16:13 Quit wodz (Quit: Leaving)
15:19:38 Join amayer [0] (
15:26:47 Quit krabador (Quit: Take the time.)
15:32:42 Quit petur (Quit: *plop*)
15:34:32 Join chrisb [0] (
16:12:14 Quit AlexP (Remote host closed the connection)
16:17:50 Join maruk1 [0] (
16:29:31gevaertswodz: I'll have some time tonight, but apart from mildly agreeing with not liking that _, I don't care a lot
16:30:09gevaertsI'll go through it, but I won't promise many interesting comments :)
16:33:13 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
16:35:53***Saving seen data "./dancer.seen"
16:37:51 Quit pamaury_ (Ping timeout: 240 seconds)
16:47:00soapwranosky74. Deep sleeper agent or hacked password?
16:49:17gevaertssoap: there's been a bunch of those, all with non-zero post counts and zero actual posts
16:51:08soapfrom 2012?
16:51:14gevaertsMy suspicion is that back when we had the huge spam wave, still in the smf 1.x days, the patches I did to alleviate the issues didn't handle post count correctly, and now that I've deleted all old bans those can be used again
16:51:23gevaertsEither that or an SMF security issue
16:52:06soapThat seems plausible. The hotmail address seemed marginal for a human to want. I was mulling over emailing it to offer a way out of the ban.
16:52:41gevaertsThat's why I asked scorche|sh to install the post count recalculation mod. Well, that and I'd like to stop not counting posts in e.g. unsupported builds, so we can actually purge old unused accounts
16:52:59gevaertsRight now 0 posts can mean "Only cares about unsupported builds"
16:53:12gevaertsAnd you need a mod to recalculate the lot :)
16:53:36soapdoes the "banned" usergroup work?
16:54:03gevaertsNot when I tried it a few weeks ago
16:54:25soapand is there a way to delete all posts by a user which is faster than clicking "remove" and then "Ok" and then "remove" and then "Ok" and then "remove" and then "Ok".......
16:54:42soapYeah, I saw that comment, was curious if anything had come of it.
16:54:56gevaertsYes, but maybe admin-only. Delete the entire account :)
16:57:05gevaertsI'm not convinced we want to allow that for every moderator though
16:58:23 Nick Guest2763 is now known as olspookishmagus (
17:08:27 Join lorenzo92 [0] (
17:33:23 Quit lorenzo92 (Ping timeout: 264 seconds)
17:38:53 Quit edhelas (Quit: Quitte)
17:47:13 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
17:48:57 Join cmhobbs [0] (
17:48:57 Quit cmhobbs (Changing host)
17:48:57 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
17:52:10 Quit maruk1 (Quit: Leaving.)
18:03:50 Join ZincAlloy [0] (
18:04:22 Quit cmhobbs (Remote host closed the connection)
18:14:06 Join the-kyle [0] (
18:26:06 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:35:54***Saving seen data "./dancer.seen"
18:39:37 Join xorly [0] (
18:52:11 Join RiD [0] (
19:06:48 Quit ZincAlloy (Quit: Leaving.)
19:10:58 Join anon [0] (
19:11:09 Nick anon is now known as Guest31769 (
19:14:41 Join lebellium [0] (
19:20:07 Quit Guest31769 (Quit: Leaving)
19:27:43 Quit xorly (Read error: Connection reset by peer)
19:28:26 Join Provel [0] (
19:28:54 Join xorly [0] (
19:30:01 Join AlexP [0] (~alex@rockbox/staff/AlexP)
19:33:24 Quit chrisb (Ping timeout: 244 seconds)
19:56:11 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
20:00:33 Quit Provel (Read error: Connection reset by peer)
20:06:42 Join MMlosh [0] (~MMlosh@2001:470:6f:23:24f:63ff:fe01:4900)
20:07:18 Quit pamaury (Remote host closed the connection)
20:27:39 Join ZincAlloy [0] (
20:30:18 Join petur [0] (~petur@rockbox/developer/petur)
20:32:11 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
20:35:56***Saving seen data "./dancer.seen"
20:42:50 Join krabador [0] (~krabador_@unaffiliated/krabador)
20:51:05 Join Rower [0] (
20:53:26scorche|shgevaerts: i'd highly suspect that these were created in the previous waves
20:53:56gevaertsscorche|sh: yes, looks likely
20:54:02scorche|shi'll look at the mod, but i really just need to redo the entire server and have a fresh installation of SMF on there
20:54:21gevaertsYes, maybe, but I bet that won't be done soon
20:54:37scorche|shand any mods need to be applied manually, but hopefully this isnt a big one
20:54:56gevaertsAnd fixing post counts would help a lot in keeping things sane I think
20:55:10*scorche|sh nods
20:55:30gevaertsTo be honest, that shouldn't have to be a mod...
20:58:36scorche|shgevaerts: can you do me a favor and email it to me, so i make sure to not forget about it this weekend?
21:01:06gevaertsFor the record, do you think purging deleted posts (*not* the archived old forums) is a bad idea?
21:04:15scorche|shnot really
21:09:45gevaertsOK. I'll ask for more opinions before I do something like that, but I know you're not around that often :)
21:13:31 Quit Rower (Quit: Hmmm...)
21:14:25 Quit kvieta (Ping timeout: 245 seconds)
21:14:40 Quit charlie (Ping timeout: 244 seconds)
21:15:51scorche|shi'm not? =(
21:16:12gevaertsYou are, but I was planning to discuss it on the forums :)
21:18:54 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
21:18:56 Join kvieta [0] (~kvieta@
21:47:09 Join charlie [0] (~c@unaffiliated/charlie)
22:05:37 Join lorenzo92 [0] (
22:07:07soapIf these accounts were made in the previous wave that would mean that the spammers are (at least occasionally) testing old accounts for viability.
22:07:28gevaertsOr maybe they never stopped
22:07:50soapYet the overwhelming majority of ban triggers never get triggered. And the overwhelming majority of those which did get triggered never made it to double digits.
22:09:07soapSince the cleanout I, obviously, have no trigger counts to point to but for a long time there I always made two bans per abusive account, a perm ban on limited metrics and a temp ban on all metrics. It was extremely rare for the temp ban to get triggered at all so I stopped.
22:09:59gevaertsYes, I think you're right
22:17:51 Quit pamaury (Ping timeout: 240 seconds)
22:18:28 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
22:22:39 Quit eeeeeta (Remote host closed the connection)
22:24:01 Join eeeeeta [0] (~eeeeeta@gateway/tor-sasl/eeeeeta)
22:34:46 Quit petur (Remote host closed the connection)
22:35:58***Saving seen data "./dancer.seen"
22:51:48 Join MarcAndersen [0] (
22:53:40 Quit amayer (Quit: Leaving)
22:57:00MarcAndersenI have an ipod video 80 gb which, when I plug it in to usb, sometimes crashes and the hard drive dies completely just like the power to it was just cut. At other times it just works and comes up as a drive. What can be wrong here?
23:04:10lorenzo92either battery is too low and the only power source is the USB, which might not overcome to the current request
23:04:42lorenzo92or the disk well, has issues....I'm no ipod experts
23:04:55lorenzo92*expert, so wait other people :D
23:12:28 Quit pamaury (Ping timeout: 244 seconds)
23:13:14 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
23:13:14MarcAndersenIt is fully charged so it is a bit strange. When it happens I have to reset it.
23:14:15 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
23:17:13 Quit fs-bluebot (Ping timeout: 256 seconds)
23:17:21 Quit bluebrother^ (Ping timeout: 240 seconds)
23:17:56 Join fs-bluebot [0] (
23:54:53 Quit lorenzo92 (Ping timeout: 265 seconds)

Previous day | Next day