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 2013-07-29

00:01:25 Quit Rower (Quit: Hmmm...)
00:03:01 Quit bertrik (Ping timeout: 240 seconds)
00:04:33 Quit lorenzo92 (Remote host closed the connection)
00:10:26 Quit pamaury (Ping timeout: 276 seconds)
00:10:43 Quit stripwax (Quit: Miranda IM! Smaller, Faster, Easier.
00:13:09coppersaratoga: iPod Classic codec performance, unboosted:
00:19:10***Saving seen data "./dancer.seen"
00:23:32 Quit lebellium (Quit: ChatZilla [Firefox 23.0/20130725195523])
00:31:26 Quit ender` (Quit: Eighty percent of all people consider themselves to be above-average drivers.)
00:35:24 Quit tchan (Quit: WeeChat 0.4.1)
01:08:51coppersaratoga: Fuze+ codec performance, unboosted:
01:11:00coppergood night
01:11:11[Saint]Huzzah. Some real results.
01:12:25[Saint]Unboosted results are a lot nicer when the targets in question very rarely, if ever, need to boost under normal circumstances.
01:13:06[Saint]...though I'm pretty sure the Classic inherits the iPod/clickwheel "Boost on UI" behavior, which is almost certainly unnecessary.
01:17:55[Saint]Good catch there.
01:18:01*[Saint] pats himself on the back
01:18:43[Saint]This might make a /slight/ difference in battery life, moreso with heavy UI interaction.
01:19:09[Saint]Its extremely likely that the Nano2G doesn't need this either.
01:19:42[Saint]Hmmm, maybe not. That's got a much lower minimum clock.
01:26:35[Saint]TheSeven: how feasible would multiple levels of boost for the Classic (and others) be do you think?
01:27:15[Saint]going from 54MHz to ~200+ on UI interaction seems more than a little insane.
01:27:24TheSeventhere are 2 ways how boost can be implemented: 1. by changing clock dividers, 2. by changing pll multipliers
01:27:57TheSeven1. only works in integer or even power-of-2 steps, 2. has several milliseconds of latency and might mess with audio playback
01:28:33TheSevenso there's an easily accessible 108MHz clock in between, but anything else between 54 and 216 might be problematic
01:29:00[Saint]also, what'd be the best way to do it? watch some thread, step up if it starts to choke, step down when its kosher again; repeat?
01:29:01TheSevenI'm wondering if unboosting makes any sense at all on this platform though
01:29:43TheSeventhese processors have very efficient sleep modes, so unless you're busy waiting for something reducing the core clock shouldn't save much power
01:29:47[Saint]Why is that? There's almost no need to boost at all.
01:30:04[Saint]gah, lag.
01:30:26TheSeventhere is a difference if you can do undervolting to an even lower level at the lower clocks
01:30:39TheSevenbut that might again have serious amounts of latency
01:31:50[Saint]Hmmm...well, perhaps there are some targets that could benefit here if I motivated myself enough to get a WIP solution implemented.
01:32:09[Saint]Our current "all or nothing" boost seems odd and inefficient to me.
01:32:31[Saint]I'm thinking something akin to an ondemand governor.
01:33:31[Saint]The guts of deciding when to boost up/down is simple - working out different clock steps for each target I may falter on, though.
01:34:42[Saint]TheSeven: re: undervolting - have you had much luck there? If so, with which values?
01:35:01[Saint]I tried very briefly, with absurdly instable results.
01:38:37TheSevenit works somewhat well on the nano2g, but it seems like there isn't much gain from lowering the clocks
01:38:53TheSevenno idea on the classic, that has been unstable as shit due to i2c race conditions
01:44:43[Saint]Playing with the Classic now without GUI boosting - no observable difference, as expected.
01:45:31[Saint]I doubt it will make any notable difference in battery life, but, needlessly boosting can't really be a "Good Thing (TM)".
01:47:10 Join r0b- [0] (~rob@unaffiliated/r0b-)
01:47:29r0b-why is the mouse disabled in Rockbox for the Sansa Clip + ?
01:47:54[Saint]I suspect this patch predates the N2G being stable (haven't looked at the date, but I recall it being around 2008 if memory serves), and definitely predates the Classic.
01:52:01[Saint]r0b-: its not just the mouse.
01:52:10[Saint]Its all HID.
01:52:35[Saint]because you haven't enabled it yet? :)
01:52:54r0b-how do i enable it
01:54:42[Saint]Sorry - that was a bit mean. That was a failed humorous attempt at saying "its not implemented for this target, and this still needs to happen - you could do so, given enough time and education".
01:54:50[Saint]Its not an on/off switch type of thing.
01:58:50SuperBrainAKyou guys know how the ipod can emulate a mouse right? is there anybody that feels up to either telling me how to make or make it emulate a mouse left click with variable frequency? (maybe use the wheel for frequency and a button to be on or off and another one for temporary)
02:00:02[Saint]teehee - the Classic defines HAVE_USB_HID_MOUSE
02:00:16[Saint]No, Classic, can't do that.
02:02:02SuperBrainAKi could just toss a 555 in my old mouse and call i it a day but the ipod would be fun :)
02:02:15[Saint]SuperBrainAK: at a guess, I'd say you want to be nosey at /firmware/usbstack/usb_hid.c
02:02:43SuperBrainAKwhat language is it?
02:03:42SuperBrainAKok thats what i know :)
02:05:28 Quit Xerion (Ping timeout: 264 seconds)
02:06:43*SuperBrainAK grabs his old laptop
02:06:51r0b-oh god :P
02:07:07r0b-dont encourage him.
02:07:45SuperBrainAKhey its running linux :) the rest run w7 or higher :P
02:07:59*r0b- will BAN u from this game!
02:08:11 Join Xerion [0] (
02:11:39 Join SuperBrainAK_ [0] (
02:13:16 Quit SuperBrainAK_ (Client Quit)
02:17:42 Quit krabador (Quit: Sto andando via)
02:19:12***Saving seen data "./dancer.seen"
02:19:17 Join krabador [0] (~krabador_@unaffiliated/krabador)
02:28:09r0b-so i have to write the HID code for the clip + ?
02:28:18 Join SuperBrainAK_ [0] (
02:29:30 Nick SuperBrainAK_ is now known as SuperBrainAKEee (
02:33:08SuperBrainAKok i have hopefully logged in as me
02:34:20[Saint]you have, yes.
02:35:11SuperBrainAKsweet, but i did accidentally paste the whole jane doe :D
02:40:20SuperBrainAKEeeso im thinking i need to be at line 325 it seems
02:53:37 Part r0b- ("Leaving")
02:56:59SuperBrainAKso what variable is it that is the scroll wheel? im assuming it is some variable controlled by another code file that outputs a "i am changing bit" and speed and direction?
03:02:02 Quit Epicanis (Ping timeout: 276 seconds)
03:23:53 Quit Zarggg (Read error: Connection reset by peer)
03:24:25 Join Zarggg [0] (
03:28:01 Part micah
03:36:42 Quit krabador (Ping timeout: 260 seconds)
04:05:08SuperBrainAKi think i might just use my teensy ++ that way i can use an old button board to a tv and have different frequencies set
04:05:34 Quit SuperBrainAKEee (Remote host closed the connection)
04:07:45 Quit derf (Read error: Connection reset by peer)
04:19:15***Saving seen data "./dancer.seen"
04:21:49 Join webguest59 [0] (
04:21:58 Quit webguest59 (Client Quit)
04:27:01 Quit amiconn (Disconnected by services)
04:27:02 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:27:07 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:27:15 Quit pixelma (Disconnected by services)
04:27:16 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:27:19 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:55:41 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
05:16:57 Join Rower [0] (
05:17:02 Join derf [0] (
05:45:23 Quit TheSeven (Disconnected by services)
05:45:32 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:58:23[Saint]Would it help anyone any if I made a generic device config template?
06:00:45[Saint]I'm sure I'm not the only one that has noticed the various device configs are in a massive mess.
06:01:55[Saint]It doesn't change a thing for existing targets, but possibly having all the configs ordered in a sane way would make it easier for those new to the code, or a specific port, to spot defines without having to grep for things they may not know exists, or read the entire config file.
06:04:30[Saint] that might just make it even easier for people to just c/p and include defines that aren't needed. Bah.
06:07:28[Saint]Stick around long enough and you'll eventually see I talk myself out of a lot of my "good ideas" without any feedback from anyone else.
06:07:39[Saint]I'm a one man debate team.
06:19:18***Saving seen data "./dancer.seen"
07:47:46 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
07:50:01 Quit tchan (Quit: WeeChat 0.4.1)
07:59:36 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
08:14:42 Quit pamaury (Ping timeout: 276 seconds)
08:19:21***Saving seen data "./dancer.seen"
08:22:38 Join ender` [0] (
08:39:52 Join tapiralec [0] (
08:40:08 Quit tapiralec (Read error: Connection reset by peer)
08:40:58 Join tapiralec [0] (
08:53:43 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:57:47 Nick SuperBrainAK is now known as DormantBrain (
08:57:51 Join habys [0] (
08:58:46 Quit tapiralec (Remote host closed the connection)
08:59:03 Join tapiralec [0] (
09:05:11 Quit rdn (Read error: Connection reset by peer)
09:13:51 Quit tertu (Ping timeout: 276 seconds)
09:19:37 Join lorenzo92 [0] (
09:21:47 Join einhirn [0] (
09:23:44 Quit bertrik (Read error: Operation timed out)
09:26:46 Join petur [0] (~petur@rockbox/developer/petur)
09:39:30 Quit bluebrother^ (Ping timeout: 246 seconds)
09:40:39 Quit fs-bluebot (Ping timeout: 264 seconds)
09:41:34 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
09:42:09 Join fs-bluebot [0] (
09:59:45[7][Saint]: have you looked at the modular design that I used in the UMSboot source code that I linked above?
10:00:29 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:11:09[Saint][7]: I haven't, no, what in particular are you referring to?
10:11:19[Saint]My phone doesn't want to give me that much scrollback.
10:19:25***Saving seen data "./dancer.seen"
10:22:42[Saint]Ooooooh - very nice.
10:22:52[Saint]Is it booting on N2G yet?
10:23:37[Saint]All my N2Gs flatly refuse to update emCORE presently, so, I'll be happy to test when you're ready.
10:24:05[Saint]I have some handrolled-by-you, non-release-version on all mine.
10:24:20[Saint][7]: ^
10:31:20[7][Saint]: yes, that's working on n2g just fine
10:31:28[7]just doesn't have an LCD driver yet, I'm working on that
10:31:38[7]once that's done I'll port it to the classic
10:32:36[Saint][7]: question for both Classic and N2G - could we jump to diagmode from the emCORE menu?
10:32:47[Saint](vastly more useful for the Classic than the N2G)
10:34:22[7]yes, that can be done
10:34:36[7]I think I removed it for flash space reasons
10:34:51[7]but you should be able to just drop it as a .ubi file into diagmode
10:34:56[7]er into UMSboot
10:35:21[7]on the nano2g it should already be present
10:36:46[Saint]Perhaps it is, maybe I'm misremembering - but I don't recall seeing it as an option.
10:37:15[Saint]As I said, though, I do have some "pretty much as soon as you got it booting" very early emCORE version on there.
10:37:19[7]it's hiding in the tools menu
10:37:39[Saint]Was this implemented at a later stage perhaps?
10:37:45[7]quite possibly
10:41:58 Join lebellium [0] (
10:48:11 Join jlbiasini [0] (~metaphysi@
10:56:47jlbiasinipamaury: ping
10:58:15pamauryjlbiasini: pong
10:59:33jlbiasinipamaury: g#525 should be ready now if it's ok you can push it after g#529 (it depends on it)
10:59:36fs-bluebotGerrit review #525 at : [Fuze+][Touchpad] Improve touchpad power managment by Jean-Louis Biasini (changes/25/525/5)
10:59:37fs-bluebotGerrit review #529 at : [RMI Driver] Implement power saving support by Jean-Louis Biasini (changes/29/529/3)
11:01:20jlbiasiniI tested it but I didn't find a way to have the debug screen showing if the down to very_low_power mechanism was working...
11:01:39jlbiasinias it is in the other loop...
11:04:49pamauryis this reporting rate code in the debug screen really needed ?
11:05:03jlbiasinipamaury: I'm also wondering about the name of touchpad_enable() if it is the function meant to be use to disable the touchpad on keylock, I would rather have it named touch_device_enable() for consistency reason with other touchscreen
11:06:29jlbiasiniwell reading the register anyway I though it was nice to report all its value, why?
11:06:47jlbiasiniactually the code also offer to change the report rate value
11:06:58pamauryno I mean, why is there some code which changes the report rate based on sensibility ?
11:08:01jlbiasiniaha no, the sensibility_counter is just a counter to avoid change from vol key to be to sensible
11:08:57jlbiasinithe patch change the old vol keys for setting sensitivity (we have a setting for that now) to vol key for either power mode or report rate
11:10:26pamauryhum, I would prefer to only change the power mode, and use a more sensible range than -15..15. For example 1 would be fully awake, 0 would be normal, -1 low power, -2 very low power, -3 sleep.
11:10:45pamaurydoesn't that seem more logical ?
11:11:59jlbiasinipamaury: yes I just used the hrd values, regarding sensibility I tested it and it seems to me enough
11:12:33pamauryAnd I know I am stubborn but I am still not convinced that changing the report rate is a win.
11:12:47pamauryBy the way: did we fix the sensibility setting range ? I don't remember...
11:13:07jlbiasini?? did it has to be fixed?
11:13:45jlbiasini25 to -25 is good for what I know
11:13:54pamauryah ok, we changed it to -25..25, good
11:15:05pamauryI've commented on the rmi patch: it's ok except for style. The other one is basically ok, except for style and 1) report rate 2) this code in the debug screen. I'll publish the comments in a minute
11:15:06jlbiasinipamaury: another thing is the low_power mode as standard, the doc actually says that on device that doesn't implement gesture like double tap & co it should be standard
11:16:17jlbiasiniregarding report rate, I think only a battery_bench can say
11:19:18pamauryah ok, well on thing is sure (that's my understanding of the spec and it can be checked using the IRQ debug screen): 1) if you don't touch the pad, the device reports nothing so report rate is irrelevant; 2) if the pad is disabled, the device reports nothing so report rater is irrelevant; 3) otherwise the screen will be on because touchpad is on and report rate costs a lot less than backlight, so it's mostly irrelevant
11:20:17pamauryOr at least, we can tweak the report rate as part of another commit if it appears that it changes the battery life, does that seem fair ?
11:21:03jlbiasinipamaury: + actually pperforming at a lower rate is not confortable on the fuze+ so lets drop that
11:21:44jlbiasinithe doc says: "This field sets the report rate for the device. It applies in common to all functions on the
11:21:44jlbiasinidevice that have a natural report rate. "
11:23:27pamauryI've published some comments, should be easy to fix
11:23:50pamauryI know i'm a bit harsh about style but we try to keep consistent style in each file, otherwise it's a mess
11:25:07jlbiasinipamaury: that's ok It's usefull for me to learn the clean way. Regarding touch_device_enable is that ok?
11:25:54pamauryI'm not decided, but you might be right for this one
11:25:58pamaurygoing for lunch, see you later
11:29:32coppersome early lunch
11:33:53*[Saint] wonders how he could reliably benchmark this slight change to the Classic
11:33:59[Saint](turning off GUI boosting)
11:34:25[Saint]I don't suspect it will make a large, or even very noticeable, difference..but, I'd like to know.
11:35:21[Saint]maybe I can use a raspberrypi, a servo, and a capacitive touchpen to twiddle with the clickwheel in a repeatable fashion?
11:35:49[Saint]that's the thing. The UI needs to be in use, and it needs to be used consistently.
11:36:09[Saint]Very hard to test this manually. And the device just sitting idle doing nothing shows me nothing. :-S
11:38:34 Quit [Saint] (Remote host closed the connection)
11:39:40 Join [Saint] [0] (~saint@rockbox/user/saint)
12:05:50 Join stoffel [0] (
12:19:26***Saving seen data "./dancer.seen"
12:20:50 Quit michaelni (Quit: Leaving)
12:29:53 Join michaelni [0] (
13:13:04 Join krabador [0] (~krabador_@unaffiliated/krabador)
13:13:35 Quit [Saint] (Remote host closed the connection)
13:13:40copper[Saint]: I'd just make the change and see if it breaks anything
13:13:50copperif it doesn't, commit it, I guess
13:14:25copperor, if it's supposed to make little difference in practice, maybe don't bother?
13:14:42 Join [Saint] [0] (~saint@rockbox/user/saint)
13:14:51coppergreat copper
13:15:09copper[Saint]: I was talking to you right after you quit
13:15:15copper11:13:41 UTC <copper> [Saint]: I'd just make the change and see if it breaks anything
13:15:18copper11:13:51 UTC <copper> if it doesn't, commit it, I guess
13:15:20copper11:14:26 UTC <copper> or, if it's supposed to make little difference in practice, maybe don't bother?
13:21:10pamauryjlbiasini: you forgot to fix the style of the if in the rmi patch ;)
13:22:02*copper is trying to remember what he wanted to ask pamaury
13:22:46jlbiasinisee my comment on the other commit
13:22:47copperah yes
13:23:07copperpamaury: is there any way to make the sdcard reader in the Fuze+ faster?
13:23:53pamaurycopper: not sure, is the read/write speed much slower compared to actual card speed ?
13:24:03copperwriting is twice slower
13:24:07copperwith my sdxc card
13:24:18pamauryhum, and reading ?
13:25:09pamauryjlbiasini: which comment ?
13:25:18pamaurycopper: can you check that too ? That would help
13:25:42jlbiasiniI made a comment but it is showed as draft
13:25:58pamauryyou need to say review and publish comment to display them
13:26:52jlbiasinimeanwhile the rmi part is rerererecorrected ;)
13:27:31pamauryjlbiasini: I have a last comment on the rmi part, I'll add it, it will take 1 minute to fix and then i'll commit
13:29:12pamaurynot let me read the comment
13:29:37pamaurycopper: how did you check the speed ? using usb or the test_disk plugin ?
13:29:49pamauryhum I don't see your comment yet
13:32:22jlbiasinipamaury: now it should be readable
13:33:15copperpamaury: dd on the sdxc device of the Fuze+ connected via USB
13:33:42jlbiasiniwhat does BM stand for?
13:33:43copperreading is over twice slower via the Fuze+, compared to the sdxc adapter in my laptop's sdcard reader
13:34:03copper8.3 MB/s vs. 19.7 MB/s
13:34:21pamauryBit Mask
13:34:51pamaurycopper: ok, do you happen to have the test_disk plugin on your fuze+ or not ?
13:36:11pamaurycan you run it to see the result ?
13:36:16jlbiasiniok so obviously I have to use them in button-fuzeplus.c too
13:36:17copperrunning it
13:36:21pamauryjlbiasini: indeed, your comment looks relevant
13:36:35pamauryjlbiasini: where is it needed ?
13:38:52copperpamaury: does test_disk produce a log file?
13:39:02pamauryyes iirc
13:40:39 Join n1s [0] (
13:40:39 Quit n1s (Changing host)
13:40:39 Join n1s [0] (~n1s@rockbox/developer/n1s)
13:41:18pamauryok so it's not usb related (I had no doubt but I just wanted to check). I'll have a look at this issue
13:41:56copperpamaury: note that I'm using a class 10 microsdxc card (64 GB), if that makes any difference
13:42:22copperwhich one is onboard flash, and which one is sdcard (A or U)?
13:43:17pamauryah damn, I didn't think about this: test_disk probably only benchmark the internal storage. A is append, U is update I think so unrelated :-/
13:44:00gevaertsAligned, unaligned, I'd expect
13:44:20pamauryyeah, more likely
13:44:59pamauryhum, ok i'll the test by myself and see what I can do about this issue
13:46:23 Join liar [0] (
13:50:41 Quit stoffel (Remote host closed the connection)
13:51:17 Join stoffel [0] (
13:53:42jlbiasinipamaury: so are all part of the comment relevant or which ones? what is where needed?
13:54:35pamaurypamaury: you pastebin patch seems ok, all the other comments of mine still apply
13:54:59pamauryI had forgotten about the vol key delay, it's much better in your patch
13:58:11[7]"note to self"? :P
13:58:19jlbiasiniyes it's because I was trying to reuse the old one where having signed value allow to add it directly to sensistivity with a division, thus avoiding a recheck on wether it was vol up or vol down
13:58:54jlbiasinithe patch is nearly ready I was waiting for your reply on the comment before going on
13:59:40pamaurygo on, update the patch, and try to take all my other comments into account ^^
14:00:17jlbiasiniyou forgot to add: "I dare you" ;D
14:11:06 Quit derf (Ping timeout: 246 seconds)
14:12:08 Join derf [0] (
14:16:54jlbiasinipamaury: done
14:19:27***Saving seen data "./dancer.seen"
14:28:17jlbiasinipamaury: there still no link on rb main page to zen? I though they could be in unstable?
14:28:25fs-bluebotBuild Server message: New build round started. Revision be72c4f, 217 builds, 18 clients.
14:29:41pamauryjlbiasini: zen is not in the trunk
14:31:19jlbiasiniI will have to add a silent function to all touch device now, if I want this touchdev disable on lock
14:32:30pamaurydid you test all the patches ?
14:32:51jlbiasiniyou mean the ones I just commited yes
14:33:36jlbiasinibut I is not possible to knox if the touchpad is actually lowering after one minute...
14:36:20 Join amayer [0] (
14:36:32fs-bluebotBuild Server message: Build round completed after 487 seconds.
14:37:14pamauryjlbiasini: it would be possible if the debug screen read the power value each time
14:39:45jlbiasinistill something to add then
14:45:52jlbiasinipamaury: well that would be tough anyway... the backlight goes down and to have it backup without the touchpad it volume key or power...
14:49:04jlbiasinipamaury: no it does works! even if the volume keys are touch as long as no activity on the touchpad occurs it set it back to the very_low_power
14:53:56jlbiasinipamaury: ok I commited a fix
14:54:10pamauryjlbiasini: you can set backlight timer to 1min+
14:54:56jlbiasinianyway it's no needed we can get the backlight by volume key
14:56:04pamauryjlbiasini: last tweak and then you are done, would be better if you can test it, you can reduce the activity timeout when testing to avoid the long wait
14:58:04jlbiasinipamaury: I have question: with that mechanism won't power state goes back to very_low_power mode after no activity occurs and the touchpad has been set to sensor_slepp :/
14:59:56pamauryhum, that's right
15:01:38pamauryI see two ways of fixing this: 1) have a boolean which hold the activation state 2) implement rmi_get_sleep_mode() and only go to very_low_power if sleep_mode is not sleep
15:03:41pamaurymaybe 1) is better
15:04:04jlbiasiniso in touchpad enable I set a boolean
15:04:48jlbiasiniand recheck it before going to very_low
15:06:21jlbiasinithat would work but not in the debug screen. Is that a problem?
15:07:51pamauryno, debug screen is for debug, use it at your own risk
15:09:20jlbiasiniregarding the change in the timeout I was thinking to have a ACTIVITY_TMO_DEBUG to use unstead of the standard one. But how can I check if I'm in the debug screen from the rmi thread?
15:11:49pamauryyou cannot really
15:15:39jlbiasinipamaury: should I add another bool debug mod?
15:17:10 Quit x56 (Ping timeout: 240 seconds)
15:17:29 Quit liar (Read error: Connection reset by peer)
15:18:29lorenzo92kugel: now i'm working on something else, BUT, how do you think it's better to move? fix the issue or simply write a simpler usb "stack" for hosted targets?
15:18:41lorenzo92of course the first would be the best but...
15:18:57 Join x56 [0] (
15:19:49lorenzo92could it be also a problem related to threading? just a guess...
15:22:51 Join liar [0] (
15:23:32pamauryjlbiasini: I'm not sure this is worth it
15:34:17 Quit habys (Ping timeout: 245 seconds)
15:36:01 Join habys [0] (
15:37:22pamauryjlbiasini: or make this activity timeout a setting :p
15:45:17 Quit amayer (Quit: Leaving)
15:45:55jlbiasinipamaury: you launch the discution about it. When you will be done i'll start implementing ;D
15:48:22pamauryIt's a half serious question. I wonder what is optimal value of this activity timeout. The valid options I see are 1) same as backlight timeout 2) some hardcoded value which work well 3) some dedicated value with a setting
15:49:46pamauryand there is 4) = any of 2) or 3) + automatically switch to very_low_power when the lcd is off
15:50:14pamaurybecause it kind of makes sense to lower the power when nothing is displayed, don't you agree ?
15:54:30pamauryjlbiasini: I have another idea, because the whole points comes down to one thing really: is the transition from very_low_power to low_power fast enough ? Suppose that it is, then if you immediately switch to low_power when the user touches the pad, it will not even notice the difference.
15:55:19pamauryI think I need to experiment by myself tonight before making a decision
15:57:56 Quit DexterLB (Read error: Connection reset by peer)
15:57:59 Join Dexter_LB [0] (
16:05:32 Quit Dexter_LB (Read error: Connection reset by peer)
16:05:38 Join DexterLB [0] (
16:05:39jlbiasinipamaury; the user setting is useless because the user will not make any difference anyway. The backlight idea seems fair
16:06:47 Join thegeek [0] (
16:06:51 Quit thegeek_ (Read error: Connection reset by peer)
16:08:10jlbiasinithe direct transition seems also a fair idea but I will have to test it during the game too to make sure it doesn't bother
16:08:46pamauryyeah, we need some tests to see if one can notice the transition or not
16:09:26jlbiasiniwhat I tested is very_low_power by itself and it wasn't good enough
16:15:43 Join tertu [0] (
16:19:29***Saving seen data "./dancer.seen"
16:27:25 Quit pamaury (Ping timeout: 240 seconds)
16:33:33 Quit n1s (Quit: Ex-Chat)
16:48:01 Join n1s [0] (~n1s@rockbox/developer/n1s)
16:57:08 Quit [Saint] (Read error: Connection reset by peer)
16:58:06 Join [Saint] [0] (~saint@rockbox/user/saint)
17:12:15 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
17:15:23 Join ikeboy [0] (
17:28:17jlbiasinipamaury: from what I can test I don't see any difference indeed... so we could set it to 5 * HZ just to avoid useless change and we are done...
17:28:51jlbiasinii tested with HZ
17:28:51 Quit ikeboy (Ping timeout: 246 seconds)
17:29:26 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:30:52jlbiasinipamaury: And if you read the golden quote you should know that we don't need extra tests... As far as I know both of us already has 2...
17:32:09 Join amayer [0] (
17:37:19 Quit krabador (Quit: Sto andando via)
17:40:19 Quit petur (Quit: Nettalk6 -
17:43:12 Join ikeboy [0] (
17:51:07[7][Saint], pamaury: the LCD driver (and related stuff) has landed, so I'd say the nano2g part is finished :)
17:51:17[7]anyone around for nano2g usb stress testing?
17:52:33[Saint]gah - unfortunately, I cannot now get to my treasure chest of devices tucked away in the closet without waking up Ms [Saint]
17:53:07[Saint]were it not 4am, this wouldn't be an issue.
17:53:24[7]well, this can wait until tomorrow :)
17:53:33[7]I have a whole ipod classic port to do...
17:53:58[Saint]*those* are right next to me...
17:54:19[7]so, what do we need for this to work on the classic?
17:54:54[7](i.e. sl8720 chipset support + ipod classic LCD support and other peculiarities)
17:56:21[7]someone interested in reading the source code, or should I go ahead and do the classic port before updating the source?
17:56:39[Saint]Is emCORE going to be pushed as the official N2G bootloader?
17:56:52[7]I see no reason to do that
17:58:54[Saint]I seem to recall the bootloader USB being completely wonky.
17:59:16[7]there's no need for bootloader USB *at all* on the N2G
17:59:17[Saint]I can't recall if it was ever fixed before N2Gs USB died completely.
17:59:36[7]well, the plan is to finally fix N2G USB
18:00:35[Saint]Hum. This is so many years ago. I may misremember this.
18:00:48[Saint]There's a chance I kludged bootloader USB in there.
18:01:34[Saint]it isn't needed, due to OF/disc mode, but - its kinda handy.
18:05:32 Join mrtux_ [0] (
18:06:06 Quit mrtux_ (Changing host)
18:06:06 Join mrtux_ [0] (~colin@unaffiliated/mrtux)
18:09:09 Quit mrtux (Ping timeout: 264 seconds)
18:09:10 Nick mrtux_ is now known as mrtux (~colin@unaffiliated/mrtux)
18:17:22 Quit ikeboy (Quit: Ex-Chat)
18:19:32***Saving seen data "./dancer.seen"
18:25:06 Quit habys (Ping timeout: 245 seconds)
18:26:00 Quit [Saint] (Remote host closed the connection)
18:26:47 Join habys [0] (
18:26:58 Join [Saint] [0] (~saint@rockbox/user/saint)
18:27:00 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:33:11 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:42:06 Join belak [0] (~belak@facebook/engineering/belak)
18:50:48 Join thegeek_ [0] (
18:54:02jlbiasiniis there ANY touchscreen device with softhold? just tried mrope500, cowond2
18:54:12 Quit thegeek (Ping timeout: 256 seconds)
18:59:26jlbiasinihey I found one!!! VX777 and the simulator doesn't compile!!
19:00:22jlbiasinikeymaps uncomplete on a plugin!!
19:00:49gevaertsThat's interesting
19:00:57gevaertsThe main build does seem to compile
19:01:34jlbiasiniit's missing TOUCHSCREEN_QUIT and TOUCHSCREEN_TOGGLE define
19:04:45*scorche|sh takes away jlbiasini's exclamation mark key
19:04:47gevaertsThat seems to fix it
19:05:14jlbiasiniI know I'm about to push the same to gerrit
19:05:53gevaertsAh, good. I won't commit then
19:06:03gevaertsPushing a button on gerrit is less work :)
19:06:31jlbiasini g#531
19:06:35fs-bluebotGerrit review #531 at : [ONDAV VX777][simulator] fix missing keymaps in test_touchscreen by Jean-Louis Biasini (changes/31/531/1)
19:07:21fs-bluebotBuild Server message: New build round started. Revision f285a0f, 217 builds, 19 clients.
19:11:12lorenzo92jlbiasini: also R1 by the way, I'm going to polish my raw patches ^^
19:11:34 Join ikeboy [0] (
19:14:33 Quit Poodlemastah_ (Read error: Connection reset by peer)
19:15:04fs-bluebotBuild Server message: Build round completed after 465 seconds.
19:15:05 Join Poodlemastah [0] (
19:16:22 Quit Rower (Quit: Hmmm...)
19:20:16 Quit mrtux (Read error: Connection reset by peer)
19:22:15 Join mrtux [0] (
19:22:17 Join Rower [0] (
19:27:29 Quit ikeboy (Remote host closed the connection)
19:36:10 Join ikeboy [0] (
19:36:57 Join krabador [0] (~krabador_@unaffiliated/krabador)
19:36:59 Quit stoffel (Read error: Operation timed out)
19:40:57 Join pretty_function [0] (~sigBART@
19:52:58 Quit krabador (Ping timeout: 260 seconds)
19:55:10[7]the ipod classic port compiled first try without warnings
19:55:14[7]now if it would just work...
19:56:15copperit always does
19:56:19copperrecently, anyway
20:05:57 Join krabador [0] (~krabador_@unaffiliated/krabador)
20:14:17 Quit pamaury (Read error: Connection reset by peer)
20:17:52 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
20:19:34***Saving seen data "./dancer.seen"
20:23:21pamauryjlbiasini: ping
20:29:46 Join einhirn [0] (
20:40:28 Quit ikeboy (Quit: Ex-Chat)
20:42:50 Quit shamus (Read error: Connection reset by peer)
20:43:16 Join shamus [0] (
20:47:33 Quit krabador (Quit: Sto andando via)
20:56:00 Quit shamus (Read error: Connection reset by peer)
20:57:18 Join shamus [0] (
21:15:30 Quit Zarggg (Ping timeout: 240 seconds)
21:22:47 Quit pretty_function (Remote host closed the connection)
21:22:52 Quit Xerion (Quit: )
21:25:45 Join rdn [0] (
21:31:53 Join evilnick_ [0] (
21:32:10 Join dokan_ [0] (
21:32:11 Quit dokan (Ping timeout: 246 seconds)
21:32:11 Quit evilnick (Ping timeout: 246 seconds)
21:33:07 Quit JdGordon (Ping timeout: 246 seconds)
21:33:08 Join JdGordon_ [0] (
21:33:08 Quit JdGordon_ (Changing host)
21:33:08 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon)
21:34:09 Join stoffel [0] (
21:37:49 Join freqmod_ [0] (
21:38:24 Quit freqmod (Ping timeout: 246 seconds)
21:43:57[7]wow, that was quicker than I had thought
21:44:08[7]now supports nano2g and classic
21:44:38[7]pamaury: ^
21:45:58 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
21:55:59 Quit lorenzo92 (Quit: ChatZilla [Firefox 22.0/20130712032019])
21:59:11 Quit mrtux (Changing host)
21:59:11 Join mrtux [0] (~colin@unaffiliated/mrtux)
21:59:54pamaury[7]: is it stable on nano2g
22:00:09[7]as far as I can tell yes
22:00:46[7]but I wasn't affected by those problems people keep reporting all the time in the first place, so it needs more testing
22:02:31[7]and, guess what, even the current rockbox ipod classic build works fine here. hm...
22:06:39[7]the nano one locks up though
22:07:43[7]are there ANY iPod Classic users here, who are having trouble accessing a current build via USB? Or trouble accessing UMSboot on Classic or Nano 2G?
22:08:58coppernot me
22:09:09coppercurrent build works fine
22:12:26amayer[7]: i havent had any problems accessing USB on my current build of iClassic
22:19:16 Quit shamus (Read error: Connection reset by peer)
22:19:38***Saving seen data "./dancer.seen"
22:20:18 Join shamus [0] (
22:20:59 Quit saratoga (Ping timeout: 250 seconds)
22:33:37 Quit y4n (Quit: PANTS OFF!)
22:36:08 Join Xerion [0] (
22:38:00jlbiasinipamaury: pong
22:40:30pamauryI have commented your patch
22:41:15pamauryWe still to experiment but you can already correct this
22:41:20pamaury*need to
22:42:58jlbiasinipamaury: I might not have time before 1 or two weeks but I'll try
22:43:11pamauryok, then i'll try
22:44:45jlbiasinipamaury: as you want actually. I'm still wanting to do it but I'll travel over ten days starting with thirsday (basically i'm going back to france across austria...
22:45:15jlbiasiniso it actually depends on the equipment od the train i'll be into
22:45:51jlbiasinieither I'll be all done saturday either i'll do it in france
22:49:02 Quit jlbiasini (Quit: jlbiasini)
23:00:50 Quit thomasjfox (Quit: Konversation terminated!)
23:02:03 Quit amayer (Quit: Leaving)
23:05:02 Quit kevku (Quit: KVIrc 4.3.1 Aria
23:10:08 Join jlbiasini [0] (~metaphysi@
23:10:38jlbiasinipamaury: well I will see what I can do now
23:14:41 Quit stoffel (Ping timeout: 245 seconds)
23:17:21 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
23:19:35 Quit kiwicam (Read error: Connection reset by peer)
23:19:49 Join kiwicam [0] (~quassel@
23:26:25 Quit n1s (Quit: Ex-Chat)
23:30:11jlbiasinipamaury: done!
23:31:53pamaurygreat, looks better :)
23:43:01 Quit liar (Ping timeout: 245 seconds)
23:55:47 Quit jlbiasini (Quit: jlbiasini)

Previous day | Next day