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 2014-03-27

00:04:56 Join nomada [0] (
00:09:54jhMikeSkugel: one issue there is that some codecs/libs expect pointers to not move (such as libmad, mod and others) between frames
00:10:43kugeljhMikeS: the codecs are halted using the existing API. i.e. when the decode loop asks for more data
00:11:28 Quit Marex (Ping timeout: 268 seconds)
00:11:33jhMikeSthey don't all work that way. libmad stores the pointer of the previous frame. things like mod grab a pointer to the buffer but are atomic and don't check it again
00:12:19 Join Marex [0] (~Marex@
00:12:28kugelI'm looking at libmad
00:12:52jhMikeSreally, I don't know all that might expect a pointer to be valid around the loop since it wasn't really a concern
00:13:05kugelafter it calls ci->get_command() it calls ci->request_buffer()
00:13:31jhMikeSatomic formats are locked in place, streaming ones are only shrinkable, not movable
00:13:53kugelwhether or not the buffer from request_buffer() is different or not is not known because the guard buffer migiht be used, no?
00:14:06jhMikeSit's a bit deeper in libamd
00:14:48jhMikeSit might be, but that only limits request size
00:14:51kugelthe stream buffers are reset afterwards
00:15:53kugelit worked for my (limited) testing too
00:15:55 Quit kuldeepdhaka (Quit: gone to fetch temporary death ;))
00:16:27jhMikeSthe concern is in the stream struct
00:16:51pamauryare there portalplayer experts around still ?
00:17:34kugeljhMikeS: I don't see it, the old pointer is lost
00:19:08 Quit ZincAlloy (Quit: Leaving.)
00:20:12kugeljhMikeS: I'm thinking this situation should be handled by all codecs, the pointer changes also when the buffer wraps around
00:20:39kugeldoesnt it? (I'm no expert in this area)
00:20:58jhMikeSpossible I misinterpreted libmad/frame.c or just forgot after so long
00:22:18jhMikeSderp stream.c resets it all
00:25:25jhMikeSand mod.c? (maybe others)
00:26:43jhMikeSone case in mpa.c is unsound
00:27:06jhMikeSthings are only reset if mad_stream_buffer is called but not if it skips it
00:27:45 Join us`0gb [0] (
00:29:22jhMikeSiirc it does that to byte-seek for a frame sync code
00:32:28kugeljhMikeS: have you looked at the patch. does it look sensible ignoring the potential codec problems?
00:32:38jhMikeSmaybe that just needs some minor tweaks
00:32:46jhMikeSso far, yes
00:33:34jhMikeSmuch as I can say right off the top of my head :)
00:33:37kugelthe idea was to halt the codec, safe the file offset, reset the buffer (and rebuffer), then restore the file offset and resume the codec
00:34:12kugelI was surprised to find that everything needed was already there so the patch is rather simple in the end
00:35:30kugelif the codec is about to request new data anyway, and buffering restores the stream properly (because of the stored file offset), then no audio samples should be lost
00:36:16 Quit GeekShad1w (Ping timeout: 240 seconds)
00:36:20kugeloh yea, during that time the PCM buffer should just be drained, and it shouldnt take long enough to go empty
00:37:07gevaertsIf all else fails, we can just make it bigger
00:37:08jhMikeSsure, most times
00:37:35jhMikeSa hard drive spinup is pretty close to the limit as long as it's full
00:39:22jhMikeS"all else" always fails :)
00:39:25***Saving seen data "./dancer.seen"
00:40:05jhMikeSso, then, atomic format (this breaks "pin handle" expectations still)
00:40:19kugelsince it's just normal rebufferin from the pcm-side it's the same problem when the audio buffer goes low
00:41:09gevaertskugel: what if the buffer is low when you do the work?
00:41:23gevaertsYou'd basically combine both bad cases then
00:41:35 Quit tertu (Ping timeout: 240 seconds)
00:42:41kugelgevaerts: buffering_reset() closes all handles (as soon as the buffering thread goes into the event loop), so if it occurs during rebuffering or just before that would be cancelled
00:44:37gevaertskugel: I mean, you're basically adding the buflib stuff time to the "regular" rebuffer time (including its spin up)
00:45:06jhMikeSwould be nice to spin the hard drive before doing the realloc
00:45:14jhMikeS*for such targets
00:45:27gevaertsSo I suspect you'd need a larger pcm buffer, which IIRC means things like DSP change latency gets longer?
00:45:42jhMikeSindeed it would
00:46:10jhMikeSthat shit really needs to not be like that but it's also sort of hard with crossfading
00:46:31kugelgevaerts: normally the DSP setting items change the pcm buffer into low-latency when in the menu
00:46:58jhMikeSbut you still initially wait for the buffer latency for the first change in a setting to be audible
00:47:31jhMikeSonce it's drained it's maintains low-latency
00:47:51kugelgevaerts: uhm, the worst case would be that it happens just after rebuffering has finished in which case the disk is still spinning
00:48:23kugelif the disk is already spinning audio decoding can be resumed quickly, it doesnt need the whole file to be buffered
00:48:55jhMikeSunless atomic format
00:50:29kugelfuck these :)
00:51:05jhMikeShaha...then I'll have to stick with an older revision myself :)
00:51:31*[Saint] doesn't think he could handle it taking any longer for DSP to catch up with the current state on some targets
00:51:34 Join Strife89 [0] (
00:51:44[Saint]For the Classic, it can be ~12s+
00:52:16jhMikeS[Saint]: really DSP should be late, with raw codec output being stored, not processed audio
00:52:47jhMikeSyou only get 12s if you're doing crossfade, it would otherwise be 3s
00:53:11 Quit pamaury (Ping timeout: 265 seconds)
00:53:13jhMikeSlow mem gets 1s
00:53:18[Saint]Don't you make me give up my crossfade, ya'hear!
00:53:26*[Saint] shakes an old man fist
00:53:29 Quit Marex (Ping timeout: 245 seconds)
00:53:33jhMikeScrossfade gets death by snu-snu
00:54:38 Join Marex [0] (~Marex@
00:55:05[Saint]crossfade on automatic track skip only is one of my favorite settings.
00:55:07 Quit kugel (Ping timeout: 265 seconds)
00:55:52[Saint]Its either that, or go through 1.5TB+ and cut out all the annoying silent leads into tracks which the mastering artist thought should be there.
00:56:00jhMikeSI never use it except when fixing it or changing code, so I could say I've used it alot, but not because I use it
00:56:37[Saint]I have a 10s fade in/out on auto-changed tracks.
00:57:01[Saint]manual change doesn;t get squat. I want it NOW!
00:57:10*gevaerts looks at [Saint] in a despairing way
00:57:14gevaertsDon't you like music?
00:57:55[Saint]Sure I do, that's why the silence at the begginning of <pretty much every track ever> needs to piss off.
00:58:30[Saint]I don;t subscribe to this mastering artist knows best business.
00:58:42*jhMikeS puts gaps at the end of the rips to avoid this problem
00:59:05[Saint]...wouldn;t that compound the issue?
00:59:23[Saint]Oh. Errr. I see. Right.
00:59:39gevaerts[Saint]: you've just basically destroyed e.g. Beethoven's fifth with those settings
00:59:41jhMikeSno, the existing gaps on the media can go after the current track or at the beginning of the next
01:00:12jhMikeSbecomes beethoven's 4.95th?
01:00:42[Saint]gevaerts: Ohhhh...what a shame.
01:00:55jhMikeSbeethoven can't hear it anyway
01:01:11[Saint]You asked if I liked *music*!
01:02:09*gevaerts hunts [Saint] down
01:02:57gevaertsAnyway, there's a *lot* of music that starts in a rather spectacular way. If you mute that, it's just not the same any more
01:03:06jhMikeSper-album audio settings maybe? :)
01:03:28gevaertsjhMikeS: this is when the album changes :)
01:04:43[Saint]I don't _mute_ it, I just blend the futzing around in with the ass end of the last track so that the lead in isn't noticeable.
01:05:03[Saint]Though, you have to keep on your toes doing that as some masters like to do this to their albums themselves.
01:05:32*gevaerts doesn't get the entire concept
01:05:45[Saint]Albums you're supposed to listen to in full, start to finish. Nuts to that.
01:05:53gevaertsWell yes
01:05:57gevaertsAlbums are a stupid idea
01:06:01gevaertsAnd so are sonfs
01:06:12gevaerts*music* is what it's about, none of your muck :)
01:06:16jhMikeSand notes
01:06:23[Saint]I *HATE* those pesky sonfs...always messing up 'ma shit.
01:06:24jhMikeSwho's got old man fist now?
01:06:51jhMikeSkids these days...
01:07:14gevaerts[Saint]: as far as I understand it, you're still lowering the volume of the starting track
01:07:41jhMikeSindeed, fade-in
01:08:08gevaertsAll I'm pointing out is that with some music, that's just a crime
01:08:12jhMikeSthough you can fade out with overlapping but just start without a fade
01:08:38[Saint]gevaerts: fortunately, its only you that listens to that music. ;)
01:09:07*gevaerts very strongly doubts that
01:09:40*jhMikeS does too, but also metal and such, sometimes back to back
01:14:02 Quit Scromple (Read error: Connection reset by peer)
01:15:42 Join GeekShadow [0] (~antoine@reactos/tester/GeekShadow)
01:17:02 Join Scromple [0] (~Simon@
01:18:29 Quit Marex (Ping timeout: 245 seconds)
01:19:54 Join Marex [0] (~Marex@
01:23:12*[Saint] wonders if RaaA could use DSP Manager
01:23:45[Saint]The fact that DSP Manager doesn't have a blacklist/whitelist makes things mightily annoying if you use DSP Manager and RaaA.
01:23:57[Saint](RaaA gets a double dose of DSP magics)
02:07:51[Saint]Damn this unstoppable force/immovable object bullshit.
02:08:18[Saint]I have hit the unavoidable wall where all the code for my theme is done, and it needs a "face".
02:08:27[Saint]I, however, am *not* an artist.
02:08:42[Saint](despite what my forum profile may or may not say)
02:12:52 Quit Scromple (*.net *.split)
02:12:52 Quit TheSeven (*.net *.split)
02:12:52 Quit bthomson___ (*.net *.split)
02:12:52 Quit scorche` (*.net *.split)
02:12:52 Quit Beta2K (*.net *.split)
02:12:53 Quit soap (*.net *.split)
02:12:53 Quit onder` (*.net *.split)
02:12:53 Quit Cultist (*.net *.split)
02:13:35 Join onder` [0] (
02:22:12 Quit |akaWolf| (Ping timeout: 268 seconds)
02:22:40RiD[Saint] why not use squares of different colors as a placeholder
02:23:57[Saint]I don't need a placeholder. Its essentially finished Barring aesthetics.
02:26:36[Saint]I do have finny little text bits like "this is a button"; "also a button"; "will do something later" already. :)
02:31:04 Join ygrek [0] (~user@
02:39:28***Saving seen data "./dancer.seen"
02:42:39 Quit onder` (Ping timeout: 245 seconds)
02:47:14 Quit ygrek (Ping timeout: 245 seconds)
02:49:26 Join Scromple [0] (~Simon@
02:49:26 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
02:49:26 Join bthomson___ [0] (sid21436@gateway/web/
02:49:26 Join scorche` [0] (~scorche@rockbox/administrator/scorche)
02:49:26 Join Beta2K [0] (
02:49:26 Join soap [0] (~soap@rockbox/staff/soap)
02:49:26 Join Cultist [0] (~CultOfThe@
02:49:55 Join onder` [0] (
02:53:52 Quit Marex (Ping timeout: 252 seconds)
02:54:44 Join Marex [0] (~Marex@
02:57:59 Join ParkerR [0] (ParkerR@unaffiliated/parkerr)
02:58:13ParkerRBought a Sansa Clip Zip
02:58:20ParkerR$32 after tax
02:59:59 Quit AlexP (Remote host closed the connection)
03:00:38[Saint]ParkerR: bit steep for a Happy Meal toy. ;)
03:01:49 Quit onder` (Ping timeout: 245 seconds)
03:02:03ParkerRFull Rockbox and Rockdoom. I'm happy :)
03:02:44[Saint]Philosophical - in part: I don't think _any_ target now can have "full" Rockbox.
03:03:18ParkerRWell looking at the themes for the original Clip it looked better than that heh
03:03:48[Saint]I was just postulating on ROMbox and other target-specific things.
03:04:08[Saint]I don't think there's a single target that makes use of the entire feature set.
03:04:18ParkerRDo all these literally look all the same or is that some image error?
03:04:35saratogathey're different images for me
03:04:54ParkerRWell different images but the same general look
03:04:59[Saint]All look different to me.
03:05:11[Saint]I guess I need a screen grab from you to see what you're seeing.
03:05:21ParkerROne step ahead :P
03:05:27[Saint]Some look very _similar_, but, all different.
03:05:28saratogayou mean they're all the same color? the clip has a fixed color screen you realize?
03:05:41ParkerRI didnt realize haha
03:05:57saratoga1 bit per pixel, colors are fixed
03:06:07[Saint]monochromatic screen.
03:06:13ParkerRAhh good to know.
03:06:30ParkerRThat explains why they were touting a 1.1" full color screen on the Zip lol
03:06:53[Saint]You'll soon get pissed off with the form factor of said screen.
03:07:02[Saint]Its about the worst shape possible.
03:07:17[Saint]Its like they tried to make a screen that would make cabbiev2 look as bad as possible.
03:07:30ParkerRHeh I like Rockboxed
03:07:33ParkerRSo it looks ok
03:07:35[Saint](and just about every other theme that isn't "fullscreen album art".
03:07:41 Quit jhMikeS (Ping timeout: 264 seconds)
03:07:50 Join ygrek [0] (~user@
03:08:43[Saint]It makes DOOM so bad that you can't see enemy sprites until they're well within "kill you easily" range. :)
03:08:56ParkerRI did notice that...
03:09:04RiDyou should be listening to music
03:09:05RiDnot playing doom
03:09:08[Saint]"Oh, that pixel in the distance is shooting at me...huh"
03:09:56[Saint]Its a cruel twist.
03:10:03[Saint]Enough RAM to load all the maps.
03:10:08[Saint]Not enough screen to see anything.
03:10:47ParkerRI still kinda like it. My iPod Touch died a little bit ago so I wanted a cheap replacement
03:11:08ParkerRAnd the Clip does what it does fairly great
03:11:19*[Saint] sees the themesite could make better use of available screen real estate
03:11:28[Saint](rather like the forums)
03:11:31ParkerR[Saint], ikr
03:12:31ParkerRThe individual theme pages stretch
03:13:14[Saint]Eeeeewwww...I get blue in your client? Ick.
03:13:50ParkerROh haha I dont do multicolored nicks :P
03:13:52[Saint]Oh. Everyone that isn't you does. I see.
03:14:22RiDwrong, i don't use multicolored nicks either
03:14:33ParkerR[Saint], you mentioned the Zip having enough RAM for the Doom maps. How do it compare hardware wise to the rest of the line?
03:14:45ParkerR*does it
03:14:55ParkerRIgnoring the screen :P
03:16:31ParkerRI see Rocbox matches up with the linux kernel stable version lol (3.13)
03:16:34 Quit RiD (Read error: Connection reset by peer)
03:35:55ParkerR[Saint], Oh also does the Zip have issues with USB storage? It doesn't seem to want to show up PC side and half the time disconnecting USB makes it panic and reboot. I've only been able to transfer files in Sandisk's firmware
03:36:42[Saint]I would love to know what we do to fuck Windows off so badly here.
03:36:45 Join onder` [0] (
03:36:54[Saint](this is Windows, yes?)
03:37:01ParkerRWindows 7
03:37:16[Saint]Color me totally unsurprised.
03:37:32[Saint]Yeah, you'll need to get used to (if you hadn't already) using the OF for USB exclusively.
03:37:42[Saint](or, get a better host OS ;))
03:37:57ParkerRAhh ok. My only other experience has been an iPod Nano 1g and the USB worked fine in rockbox
03:38:34[Saint]All the Clip variants are affected AFAIK.
03:38:45ParkerR[Saint], Arch on my small laptop, Windows on my laptop turned desktop :)
03:39:29[Saint]That's...quite a contrast. How do you not shoot yourself in the fact every time you transition from sanity to...Windows?
03:40:05ParkerR:P well I do have my Windows themed a lot like I do linux and have cygwin
03:40:09ParkerREases the pain slightly
03:54:03ParkerR[Saint], where is the OF music? I don't see it when mounted as mass storage
03:56:04[Saint]MSC/MTP weirdness.
03:56:17ParkerRAha it has to be in MTP
03:57:12ParkerRMTP on the Nexus 7 left a bad taste in my mouth
03:57:20ParkerRNever liked it
04:08:22 Quit pixelma (Disconnected by services)
04:08:22 Quit amiconn (Disconnected by services)
04:08:22 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:08:22 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:08:24 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:08:24 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:08:26[Saint]JdGordon_: (just thinking about stuff) - how hard/easy would it be to decouple the image offset from the preload call?
04:08:50[Saint](probably a bit vague)
04:08:59Strife89ParkerR: $30 for a ClipZ? I jelly.
04:09:18[Saint]I was thinking in some cases it maght make more sense to set the offset at the time the image is called.
04:09:30[Saint]Rather than when it is preloaded.
04:09:42ParkerRStrife89, Bestbuy had if for what I thought was $34. Rang it up and it ended up being $29
04:10:08[Saint]The benefit here is that it would be possible to have all theme elements in a single bitmap separated only by their offset.
04:10:09Strife89A local sale?
04:10:32ParkerRStrife89, well judging by it's $30 on Amazon its decent
04:10:45ParkerRIt was on their "Clearance" rack
04:10:52JdGordon_[Saint]: yeah, I think thats a bad idea
04:11:04JdGordon_it would make very messy code to read for the displaying
04:11:41[Saint]I hate that if you want to display the same image at different offsets it needs to be proloaded twice.
04:11:57[Saint](and we don't seem to be able to actually tell that its the same image, and *do* preload twice...)
04:12:36[Saint]It would be marginally better if there was de-dup in bufflib.
04:14:08[Saint]With my way of thinking a theme could consist of two images. A backdrop (if required) and a single bitmap comprising all other elements.
04:14:18ParkerRStrife89, And considering I don't have Prime, it was nice gettings it for slightly cheaper and instant gratification
04:14:27JdGordon_[Saint]: wait, what does offsets mean here?
04:14:34JdGordon_is it the offset into the bmp?
04:15:10JdGordon_yeah, apart from image strips I don't tihnk supporting anything like tha is a good idea for readability
04:15:52[Saint]I guess, then, I should be poking kugel for deduplication.
04:16:27[Saint]So that preloading 10 of the exact same images just to get unique offsets doesn't *actually* preload the same image 10 times.
04:16:41JdGordon_well, something like that could be done
04:16:50JdGordon_I'd be interested in an actual use case though
04:19:51JdGordon_If all you care about is not loading the same image file twice it could (potentially) be dead simple
04:20:08[Saint]Well. Lets ignore my use case, and offsets entirely. And focus on the (probable) situation where a user might want to preload the same image twice with different idents to handle a conditional state.
04:20:14[Saint]Yeah. Right.
04:20:18JdGordon_you'd still have multiple %xl(id, filename) lines with the same filename but that doesnt mean more space
04:20:25*[Saint] nods
04:20:37JdGordon_Why would you do that though?
04:21:31JdGordon_ignoring offsets, if you're preloading the same image more than once you're being a dill
04:22:21JdGordon_it would actually be *really* simple to do what you want
04:22:32[Saint]right. Ugh. I fucked that up. No. *not* ignoring offsets, that's kinda critical to my argument. :)
04:23:44[Saint]I understand about readability, but really, having the image offset coupled to the preload cal is balls.
04:24:22[Saint]I should be able to preload the image and then display it at any given offset with the draw call.
04:24:42[Saint]Not stack up as many preloads as I need offsets.
04:24:43JdGordon_in load_skin_bitmaps(), in the else{} case: 1) first check if img->loaded != true; 2) copy the filename before load_skin_bmp() overwrites it, 3) load_skin_bmp(), 4) loop through the remiander of the list and copy the buflib_handle to all the img's with the same filename
04:25:24JdGordon_[Saint]: I understand what youre saying, but imagine how horrible the display lines would look if you added 2 more params to each %xd()
04:25:33JdGordon_much cleaner to put that in the load
04:25:40[Saint]Ideally, all the proload call should care about is how many subimages the image has.
04:25:47[Saint]...actually. Not even then.
04:26:03JdGordon_I disagree
04:26:29[Saint]%xd(foo, x,y) doesn't seem to bad to me.
04:26:37[Saint](viewport relative, of course)
04:26:47JdGordon_the %xl() is literally "use this ID to draw <image filename> [from x, y offsets] when I ask for it
04:26:55[Saint]go to 0,0 if its left out.
04:27:19JdGordon_it would also be ambiguous
04:27:36ParkerRShe's a beaut' :)
04:27:38JdGordon_%xd(a, 3, 2) == image a, offset (into the strip) 3 item 2
04:28:22[Saint]I'll say again. WRT readability, I *do* understand...but, c'mon, its not like its human readable as-is. And with the fact that users can freely mix old and new syntax readability pretty much went out the window.
04:29:08[Saint]The code is human the author.
04:29:16[Saint](sometimes not even then)
04:29:18JdGordon_that's a whole other problem :)
04:29:25JdGordon_but why should we try to make it worse?
04:29:41JdGordon_but my understanding is you want to make sure the same *image file* isnt loaded twice?
04:29:44[Saint]Personally, I don't think it is. But...yadda yadda, subjective.
04:30:55JdGordon_what about this argument then: with my suggestion changing the size (or position, or file) of an image would be a 1 line change, with yours it needs to be changed for every identifer use
04:30:58[Saint]Basically. Yes. If I can't manipulate the offset from the display call, I want to make sure that I'm not crippled if I preload the same image umpteen dozen times at different offsets.
04:31:40[Saint]The latter is acceptable. Though I personally believe the former is ideal, albeit disruptive.
04:31:57 Join ygrek_ [0] (~user@
04:32:10JdGordon_IIRC the whole .bmp is loaded, so try adding a few %xl() lines with the same filename and different id
04:32:27JdGordon_ignore ram usage, it will jus work
04:32:44JdGordon_with a few lines of code change that won't use any more RAM (well negligable amount)
04:34:57 Quit ygrek (Ping timeout: 240 seconds)
04:38:34JdGordon_[Saint]: consider this wps
04:38:50JdGordon_which would you rather add or remove an image from?
04:39:03JdGordon_or change the image dimensions
04:39:29***Saving seen data "./dancer.seen"
04:40:15 Join Zarggg [0] (
04:47:18 Quit us`0gb (Ping timeout: 264 seconds)
04:47:25 Join us^0gb [0] (
04:59:32JdGordon_[Saint]: ^ ?
05:01:46ParkerRRockboxed is bext WPS :P
05:02:32 Join amithkk [0] (uid4289@2buntu/writers/amithkk)
05:02:36[Saint]JdGordon_: the latter wouldn't bother me.
05:03:15JdGordon_thats what I'm proposing
05:03:51[Saint]Crap. Former. *former*.
05:03:59JdGordon_no backsies!
05:16:45ParkerR[Saint], I'm not sure how easy this would be to doo but: Instead of expanding the theme listing to fit width you could let the user choose their device and then when they hover on a skin have it pop up the device and skin image sorta like a little preview
05:22:20 Quit TheSeven (Disconnected by services)
05:22:36 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:24:37 Quit Marex (Ping timeout: 240 seconds)
05:26:52ParkerR[Saint], shitty mspaint mockup
05:32:39 Join Marex [0] (~Marex@
05:33:13saratogawe have the graphics for the various players, so probably not too hard to do, but we as a project don't do much web stuff
05:34:15[Saint]The general look and feel of the forum and theme site should say as much to the trained observer.
05:42:32ParkerR[Saint], I have about red blob left of ammo
06:03:20ParkerRIt's weird having an FM radio and even weirder that I can actually record it. Heh kinda nice.
06:13:36 Join ygrek [0] (~user@
06:13:57 Quit ygrek_ (Ping timeout: 240 seconds)
06:16:11ParkerRIs there any microphone decibel meter for rockbox?
06:16:59 Join tertu [0] (
06:23:49[Saint]ParkerR: I believe the recording screen has a rec level IFF the screen is large enoguh.
06:23:55 Join goom [0] (
06:24:04[Saint](its hardcoded, so it may push it offscreen)
06:24:29 Part goom
06:24:43[Saint]proper rec screen skinning support has been a want of mine for a while now.
06:33:13[Saint]I probably should check that for you...but, I'm in a lazy mood so speculation is all you get for now. :)
06:35:08ParkerRHeh I'd go check but its in the kitchen charging and I dont feel like getting up
06:35:16ParkerR[Saint], high five for lazy :D
06:38:50 Join DTSyX [0] (
06:39:32***Saving seen data "./dancer.seen"
06:43:01 Quit DTSyX (Client Quit)
06:47:44 Quit x56 (Ping timeout: 252 seconds)
06:53:18 Quit tertu (Ping timeout: 265 seconds)
06:56:12 Quit ygrek (Ping timeout: 265 seconds)
07:08:14 Join |akaWolf| [0] (~akaWolf@
07:14:11ParkerR[Saint] et whoever: Enjoy a beer on me ;) Donated $5
07:24:01 Quit amithkk (Quit: Connection closed for inactivity)
07:27:03 Join x56 [0] (
07:33:58 Quit x56 (Quit: Peace)
07:44:07 Quit Zarggg (Ping timeout: 268 seconds)
07:44:41ParkerR[Saint], I just now realized this has an OLED screen after reading the rockbox wiki lol
07:45:07 Quit amiconn (Remote host closed the connection)
07:45:07 Quit pixelma (Remote host closed the connection)
07:45:19[Saint] read the manual?!?
07:45:37[Saint]Oh. Wiki. Whoops. I misread.
07:45:54[Saint]Still, you just put yourself far ahead of the pack of most users. :)
07:46:00 Join pixelma [0] (pixelma@rockbox/staff/pixelma)
07:46:01 Join amiconn [0] (amiconn@rockbox/developer/amiconn)
07:47:03ParkerR[Saint], heh I tend to read up when bored
07:47:13 Join x56 [0] (
07:47:45[Saint]Polite. Courteous. Reads the documentation. Donated to the project.
07:47:50ParkerR[Saint], now to find a nice dark theme :D
07:47:52[Saint]Am I being trolled?
07:47:57ParkerRNope :D
07:48:17 Quit JdGordon_ (Ping timeout: 240 seconds)
07:48:34[Saint]Find a qualified member of the opposite sex and breed.
07:48:36ParkerRI've been around rockbox since my old blue iPod Mini
07:48:39[Saint]We need more of you.
07:49:05ParkerRDoom in greyscale was fun
07:49:29[Saint]The mini is a fantastic target.
07:49:40[Saint]If the iPod Classic had a greyscale LCD I would be in heaven.
07:49:46ParkerROh god, all the ipodlinux stuff for it too
07:50:26ParkerRThose were great days
07:50:31copper06:47:46 UTC <[Saint]> Polite. Courteous. Reads the documentation. Donated to the project.
07:50:33 Join fragilematter [0] (~fragilema@unaffiliated/fragilematter)
07:50:34copperIT'S A TRAP!
07:50:54[Saint]My thoughts exactly.
07:51:21copperhe's trying to seduce you in order to get you to do things!
07:51:55ParkerRNever... muhahaha
07:52:13copper[Saint]: also,
07:52:34ParkerRI loved that theme
07:52:38ParkerRSo freaking simple
07:53:00copperit's available for 320x240 targets (iPod Classic / Video)
07:53:10ParkerRThen we got color and had to make them look pretty
07:53:13[Saint]ParkerR: <trolling> regarding "dark themes", its an OLED, just turn the contrast (I think we call it brightness for some reason, I don't think we differentiate) down. </trolling>
07:53:55ParkerRHeh yeah what lead me to reading up was when I turned "Brightness" down it looked an awful lot like contrast adjustment
07:54:15[Saint]copper: while that is very cute, unfortunately it can't replicate the only actual good thing about greyscale LCDs.
07:54:21copperyeah :-
07:54:24[Saint]visible in full sunlight without a backlight.
07:54:40ParkerRDarker areas casting the shadow on the screen (not quite the right workd for that)
07:55:09[Saint]copper: nice work color matching, though.
07:55:20[Saint]Very, very close.
07:56:18copperI can't tell because I don't have an actual grayscale iPod to compare
07:56:36ParkerRI think I still have the entire factory manual for my iPod Mini that I primted off forever ago
07:56:55copperI've wanted an iPod since it was first introduced in 2001, but I held off for 11 years!
07:57:10[Saint]Good thing too.
07:57:40[Saint]Barring the Mini, Color/Photo/4G and Video, they were all pretty shit.
07:59:05copperyou just cited half the lineup :D
08:00:02[Saint]Oh. And the Nano1G, of course, but that's just a tiny flash Video.
08:00:36 Join ygrek [0] (~user@
08:00:37[Saint]n2G is rather powerful but sounds like ass.
08:01:19 Join DTSyX [0] (
08:01:49 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
08:02:02ParkerRThe green trapper-keeper just makes this all that much more legit
08:02:50 Part DTSyX
08:03:34ParkerRAlso the iPod superdock I never built
08:03:41ParkerRBack when Engadget was good...
08:03:53ParkerRThat entire guide was on Engadget
08:04:56ParkerRApril 18th 2006
08:05:20ParkerRToo many memories!
08:06:07ParkerRPaper shows that I printed it on March 14, 2008
08:09:26 Quit [Saint] (Remote host closed the connection)
08:10:49ParkerRBye bye...
08:11:09 Join [Saint] [0] (~saint@rockbox/staff/saint)
08:16:11JdGordon[Saint]: wanna give g#776 a try?
08:16:13fs-bluebotGerrit review #776 at : don't load the same bmp image more than once in a skin by Jonathan Gordon
08:17:16ParkerR[Saint], this is actually my first device with an OLED screen. :D
08:21:21ParkerROk I lied... Pulse smartpen
08:21:28ParkerRBut that's the only two
08:21:49ParkerRIt is my first full color OLED though heh
08:24:08 Join ender` [0] (
08:25:39 Join kugel [0] (~kugel@rockbox/developer/kugel)
08:26:00ParkerRWoever takes the time to make these PDF manuals: thank you. I'm amazed at the detail
08:30:59[Saint]It all derived from a single source with clever conditional trickery.
08:31:47ParkerRBase example then modify slightly to match player? :P
08:32:05[Saint]...not to diminish the effort that went into it, I mean, its just not quite as large an undertaking as it seems.
08:32:17ParkerRYeah I get that
08:32:32ParkerRStill very helpful
08:32:33[Saint]We don't separately maintain 30 pdfs. Nuts to that. :)
08:33:16ParkerR[Saint], Done anything with your RaspPi as of late?
08:33:52[Saint]Nothing relevant to Rockbox.
08:34:20ParkerRWell I get that. :P I remembered also talking to you a little bit in #raspberrypi
08:34:37ygrekwhy doesn't rockbox on clipzip have single key keylock?
08:34:49ParkerRAlthough thanks for the idea... portable screen plus the SDL build on the Pi heh
08:35:18ygrekis it possible to easily hack the power button to be a keylock in keymap?
08:35:30ygreke.g. in player mode?
08:35:50ParkerRWell short press of power already is a back up/exit
08:35:51[Saint]ygrek: not easily, no.
08:36:03ParkerRAnd long press is power off
08:36:16[Saint]hence the reason why it requires 2 keys to begin with.
08:36:18ygrekParkerR, it is not very needed in player mode - agree?
08:36:34[Saint]the key maps are VERY crowded.
08:36:34ParkerRI couldn't even get the two key to work. Pressing power and select didn;t seem to do anything
08:36:38ygrekyes, I understand that there are not enough keys :)
08:37:18[Saint]Which is one if the reasons why softlock takes two keys to begin with.
08:37:28[Saint]*one of
08:38:12[Saint]This edit would be possible, but extremely non-trivial to implement consistently.
08:38:46 Join DTSyX [0] (
08:39:02ParkerROh I was thinking "home" was short power press haha
08:39:23[Saint]And we wont have different key(s) doing different things in different screens unless its absolutely unavoidable.
08:39:35***Saving seen data "./dancer.seen"
08:39:44ygrekI see..
08:39:46ParkerRWoo double negatives!
08:39:50ygrekis there support for double click?
08:40:21[Saint]Just long/short press.
08:41:03ygrekall other buttons long presses also occupied?
08:41:11ygrekthere should be a way out!
08:41:27[Saint]yes. they are.
08:41:33 Quit DTSyX (Client Quit)
08:42:41[Saint]The current implementation makes it very unlikely that softlock will be accidentally engaged or disengaged.
08:43:02[Saint]Which is a huge disadvantage of having it as a single key.
08:43:02ygrekand makes it very unlikely to press with one hand..
08:43:17[Saint]That's part of the point.
08:43:50[Saint]If it can be disabled easily, it is useless as a soft lock.
08:43:54ygrekI believe using the power button for navigation was a bad choice from the beginning
08:44:22[Saint]We don't care about what the original firmware does.
08:44:33[Saint]or how keys are labeled.
08:44:49ygrekjust that the power button is not easily pressed
08:44:56pixelma[Saint]: I use the same image at different spots without preloading more than once since... aaages in my theme. Just set up different viewports, %xd is viewport relative
08:45:00[Saint]We try to use a consistent key map across all our devices.
08:45:57[Saint]So what the key is _supposed_ to do isn't necessarily what it will do in Rockbox.
08:46:16ygrekyeah, not a nice UI if you ask me, but nevermind :)
08:46:25 Join newbie|2 [0] (
08:46:28ygrekcan you please point me how is long press handled
08:46:35ygrekI mean can't seem to find it in keymap
08:47:15[Saint]button repeat.
08:48:37ygrekhome+repeat is used for quickscreen
08:48:47ygrekI wonder if I can change that to be a keylock..
08:49:13ygrekfor a personal hack
08:49:21[Saint]You could, yes.
08:49:31ygrekbtw, what about having a keylock as a menu item in that quickscreen?
08:49:44ygreklike a general solution?
08:50:02ygrekor is it not really the problem here?
08:50:07ParkerRKinda remonds me of those old VFDs
08:50:11[Saint]We would need to change the quickscreen in that case.
08:50:24 Quit newbie|2 (Client Quit)
08:50:30[Saint]Or, add a (pretty much pointless) soft lock menu item.
08:50:43[Saint]Quickscreen only displays static menu items and settings.
08:50:54ygrekwhy pointless?
08:51:10[Saint]Because no one wants to dig through a menu to find a soft lock.
08:51:12ygrekif the blind predefined seq on actions leads to keylock - it is ok
08:51:21ygrekif it is only two presses - it is fine
08:51:38[Saint]We are *very* reluctant to add new menu items.
08:51:40ygrekquickscreen is callable from anywhere with one long home press?
08:51:59[Saint]To add a new menu item it needs a justifiable use case.
08:52:03ygrekthat's understandable, considering how crowded is it
08:52:05 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:52:06[Saint]And this (sorry) doesn't. (IMO)
08:52:33ygrekso that's the question - do people find the current keylock "good enough"?
08:52:41ParkerRWhat exactly is being discussed? The Home + Select locks seems fine here
08:52:43 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:52:49ygrekok, will try a personal hack for now
08:53:01[Saint]The only we we have to measure "good enough" is people not complaining.
08:53:04ygrekParkerR, the home + select is not doable with one hand -> baaaad
08:53:07[Saint]And, they don't. SO...I guess?
08:53:12ygrekI complain! :)
08:53:29[Saint]Yes. But you are one in a rather large number. :)
08:53:59ygrekhow one gathers a majority here? bribing, lobbying, any other approved methods?
08:54:14[Saint]With edge cases like this unless we really do have a gaping hole in the functionality we'll usually point to the sources and the licensing.
08:55:47*ygrek agrees and saddens
08:56:25[Saint]Allowing users to map out custom keys is a rather sensitive issue here.
08:56:44[Saint]There really is no way to do it nicely that won't potentially end in a support nightmare for us.
08:57:26ParkerRygrek, I don't know about you...
08:57:27ygrekconsidering the specifics of the device (7 keys) - looks like
08:57:51ygrekParkerR, what theme is that (old VFD)
08:58:34ygrekParkerR, looks rather convoluted!
08:58:43ParkerRTrue! :D
08:59:12ParkerRygrek, I agree with you that a one handed would be nice but more with [Saint] on the issues with implementing it
08:59:38[Saint]Being able to operate it so easily defeats the purpose.
09:00:07[Saint]Its _supposed_ to be slightly difficult to (dis)engage. :)
09:00:09ygrekthat's why all phones out there come with a keylock that requires three hands to unlock it (sarcasm)
09:00:37ygrekbtw old nokia dumbphones have a keylock with two keypresses
09:00:42ygrekfrom the quickscreen
09:00:49ygrekor at least I configured it so iirc
09:01:37ygrekactually 3 keypresses - get to the main screen, invoke context menu, click on the first menu item
09:01:45ygrekunlock was two keypresses
09:01:53[Saint]This debate (or the root of it, rather: end-user customizable keymaps) comes up maybe once or twice a year.
09:02:11ygrekI didn't mean user-customizable
09:02:14[Saint]And the usual consensus ends up being that there just plain isn't a nice way to safely implement it.
09:02:46ygrekstill, what about an option in a quickscreen?
09:03:14*ParkerR points to the sources :P
09:03:17ygrekkeylock -> one key press to get to QS, one click (if it is a first item) to lock the keys
09:03:26ygreki.e. you will accept such patch?
09:03:26[Saint]That would mean we would have to change the way the quickscreen behaves. But that may not be a bad thing.
09:04:06 Quit Marex (Ping timeout: 264 seconds)
09:04:15[Saint]Currently the quickscreen will only accept settings, which this isn't.
09:04:53ygrekok, will see how much hacking energy I will have on this one, maybe with some luck it materializes in something
09:04:55 Join Marex [0] (~Marex@
09:06:18[Saint]The problem would be disabling it again.
09:06:37[Saint]Enabling it with the quickscreen...yeah, I can maybe see that.
09:06:41[Saint]What about disabling it?
09:07:14ygrekI don't how the keylock is handled inside
09:08:13ygrekthat depends whether it is possible to hook at the point when keylock actually prevents key presses
09:08:20ygrekI dunno
09:08:42ParkerRKeys lock, can't unlock haha. That would be quite the issue
09:10:42 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
09:11:22 Join maruk1 [0] (
09:16:38 Join petur [0] (5bb7304d@rockbox/developer/petur)
09:17:45 Quit [Saint] (Remote host closed the connection)
09:18:54 Join [Saint] [0] (~saint@rockbox/staff/saint)
09:19:50ParkerRThe wiki mentions a diagnostic mode on the OF of the Clip Zip. Is this user accesable? Kinda want to see it for shits. (Kinda like digging into the iPod diagnotic tests)
09:25:16 Quit onder` (Ping timeout: 240 seconds)
09:25:20 Join onder`_ [0] (
09:25:38 Nick onder`_ is now known as onder` (
09:30:56 Quit bertrik (Ping timeout: 240 seconds)
09:35:42 Quit ygrek (Ping timeout: 265 seconds)
10:06:50 Quit mc2739 (Ping timeout: 245 seconds)
10:07:07 Quit kuldeepdhaka (Ping timeout: 265 seconds)
10:09:00 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
10:09:18 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
10:09:23 Join ygrek [0] (~user@
10:26:26 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:39:38***Saving seen data "./dancer.seen"
10:51:22 Quit pamaury (Quit: No Ping reply in 180 seconds.)
10:51:31 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:56:31 Join ikeboy [0] (
11:47:36 Quit bluebrother (Disconnected by services)
11:47:41 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
11:48:55 Quit fs-bluebot (Ping timeout: 245 seconds)
11:50:22 Join fs-bluebot [0] (
11:52:36 Quit ikeboy (Quit: Leaving)
12:20:27 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
12:39:29 Join cmhobbs [0] (
12:39:40 Quit cmhobbs (Changing host)
12:39:41 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
12:39:42***Saving seen data "./dancer.seen"
13:07:36 Quit kugel (Quit: Lost terminal)
13:16:16 Quit Marex (Ping timeout: 240 seconds)
13:16:57 Join Marex [0] (~Marex@
13:41:44 Quit cmhobbs (Ping timeout: 255 seconds)
13:44:14 Quit ygrek (Ping timeout: 268 seconds)
13:50:17 Join amayer [0] (
13:50:40 Quit amayer (Read error: Connection reset by peer)
13:58:39 Quit Marex (Ping timeout: 252 seconds)
13:58:50 Join einhirn [0] (~Miranda@2001:638:605:4:4c9f:7078:3d7a:b56d)
14:00:01 Join Marex [0] (~Marex@
14:08:38 Join amayer [0] (
14:09:01 Quit amayer (Read error: Connection reset by peer)
14:25:28 Join amayer [0] (
14:28:18 Quit alucryd (Quit: Alucryd out.)
14:28:49 Join ygrek [0] (~user@
14:36:24 Join alucryd [0] (
14:36:24 Quit alucryd (Changing host)
14:36:24 Join alucryd [0] (~alucryd@archlinux/trusteduser/alucryd)
14:38:44 Quit Marex (Remote host closed the connection)
14:39:03 Join Marex [0] (~Marex@
14:39:43***Saving seen data "./dancer.seen"
15:04:13 Quit amayer (Quit: Leaving)
15:05:10 Join amayer [0] (
15:08:20 Join amayer_ [0] (
15:10:15 Quit amayer_ (Client Quit)
15:10:28 Join amayer_ [0] (
15:11:11 Join amayer__ [0] (
15:12:09 Quit amayer__ (Client Quit)
15:12:09 Quit amayer_ (Client Quit)
15:12:22 Join amayer_ [0] (
15:13:25 Quit amayer_ (Client Quit)
15:13:34 Join amayer_ [0] (
15:18:37 Quit Strife89 (Ping timeout: 240 seconds)
15:23:06 Quit amayer_ (Remote host closed the connection)
15:31:26 Quit kuldeepdhaka (Ping timeout: 265 seconds)
15:34:07 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
16:05:07 Quit mc2739 (Read error: Connection reset by peer)
16:06:01 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:19:39 Quit the-kyle (Remote host closed the connection)
16:28:03 Join the-kyle [0] (
16:29:30 Join krnlyng [0] (~liar@
16:39:44***Saving seen data "./dancer.seen"
16:40:34 Join ikeboy [0] (
16:42:56 Quit us^0gb (Quit:
16:45:36 Quit krnlyng (Remote host closed the connection)
16:47:26 Join krnlyng [0] (~liar@
16:50:36 Quit amayer (Quit: Leaving)
16:51:57 Quit kuldeepdhaka (Ping timeout: 265 seconds)
16:52:12 Join Strife89 [0] (~Strife89@2602:306:bce1:8c20:395b:9c9b:c5fa:86af)
16:56:43 Quit ikeboy (Quit: Leaving)
16:58:51 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
17:04:40 Quit Wardo (Quit: Blarglarg)
17:05:54 Join rela [0] (~x@pdpc/supporter/active/rela)
17:09:12 Join RiD [0] (RiD@
17:25:04 Quit rela (Read error: Connection reset by peer)
17:33:04pamaurygevaerts: are there still pp developers in the project ? I mean people who know the not-written-anywhere details of the soc
17:33:26gevaertspamaury: Try waking up amiconn
17:33:39gevaertsHe's not very active these days, but he is in the channel :)
17:36:20pamauryamiconn: ping
17:37:26pamaurydamn, the portplayer platform is a mess ^^
17:39:02 Quit Strife89 (Ping timeout: 265 seconds)
17:41:02 Quit petur (Ping timeout: 245 seconds)
17:46:29 Join AlexP [0] (~alex@rockbox/staff/AlexP)
17:47:55 Quit Zagor (Quit: Clint excited)
17:55:29 Quit the-kyle (Remote host closed the connection)
17:56:57 Join Strife89 [0] (~Strife89@2602:306:bce1:8c20:442c:50cc:c691:9d4f)
18:16:42 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:25:35 Quit krnlyng (Remote host closed the connection)
18:26:28 Join bertrik [0] (
18:26:28 Quit bertrik (Changing host)
18:26:28 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:39:48***Saving seen data "./dancer.seen"
18:41:10jhMikeSpamaury: I could possibly answer something, depending on the target and which area
18:42:12 Quit maruk1 (Quit: Leaving.)
18:47:03pamauryjhMikeS: I would like background knowledge on the portalplayers, like where does the information we have comes from and then some more technical points like are the cpu and cop running at the same frequency and things like that
18:47:34pamauryI'm working on the sansa view which uses "pp6110" (it's called GoForce but it's very similare to pp5522)
18:47:52jhMikeSAlot from iPod Linux initially and more reverse engineering after that
18:47:57jhMikeSMaybe other sources
18:48:01jhMikeSI don't know
18:48:11jhMikeSBut the cores run at the same frequency, yes
18:48:44saratogayou're dusting off the View port after all these years?
18:49:03pamaurysomeone offered me one so yeah I'm having a look
18:49:16 Quit kuldeepdhaka (Ping timeout: 240 seconds)
18:49:30pamaurywhy didn't the View port work out ?
18:50:13jhMikeSThere was stuff from Mr. H
18:50:16saratogaeverything PP is undocumented, and was reverse engineered at great difficulty
18:50:41saratogathe 6000 series eventually became tegra, and was only ever used in one or two devices, so no one was enthusiastic about it
18:51:10saratogaIIRC e200tool was half working on it (code execution but no ability to talk to NAND so you couldn't really do anything)
18:51:30saratogathen sandisk dropped PP in favor of AMS
18:52:32 Quit ygrek (Remote host closed the connection)
18:52:51 Join ygrek [0] (~user@
18:53:45 Join the-kyle [0] (
18:55:46pamauryI see
18:58:08pamauryI'll see where it goes :)
18:58:09saratoganot saying you shouldn't go for it, just saying it may not be good for your sanity :)
18:58:31pamauryI think I've lost all sanity at RE a long time ago ^^
18:58:49saratogayou should finish that sansa connect port
18:59:01saratogai think its almost all working except a way to install without modding the hardware
19:00:57pamauryisn't it hard to find ?
19:01:55saratogaah no idea
19:02:37pamauryand what are the chances I find some installl method ?
19:02:48saratogayeah i'm just kidding
19:03:21saratogaare you low on devices these days/
19:03:26 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka)
19:03:52copperhe's baby sitting my Fuze+ :P
19:04:15pamauryno, I just need something else because always working on imx233 is kind of boring and frustrating
19:04:19saratogawas going to say i can probably dig up some hardware that i won't have time for
19:04:27saratogaso you want PP?????
19:04:52saratogaimx was too documented and not buggy enough
19:05:36saratogai want something where the cache coherency doesn't work and different memory addresses have different access speeds because memory controllers are hard
19:05:41pamauryit's buggy enough ^^ I have zillions of tcc based players but they all use NAND which is a pain to handle
19:06:16saratogai think funman has my old Nano 2G if you want to play around with that
19:06:33pamaurynot so sure, I'll stick with my players for the moment
19:06:43saratogacould probably also find you a classic 6G (no nand at all!)
19:06:55pamauryI still have much work to finish: finish rbutil install for some imx233 devices, and finish FTL RE too
19:07:15saratogawhat about just replacing the FTL with the Nano 2g one?
19:07:55pamaurylack of interoperability
19:08:02saratogawith the OF?
19:08:20pamauryand basically if you break the FTL on those devices, there are high chances the OF won't even boot so then it's a pain to flash a new bootloader
19:09:04saratogadon't they have a nice failsafe USB mode?
19:12:14pamauryyeah but that doesn't help because 1) you need special tool to upload code 2) you need a special build to upload to be able to read/write the firmware. So it's not exactly user friendly to update the bootloader this way.
19:12:27pamauryThe OF provides vendor SCSI commands to do that which is much nicer
19:12:41pamauryof course we could implement them but then that's more work
19:14:37saratogaah ok, i assumed they would still work with a broken NAND
19:14:57pamauryunfortunately not
19:26:07 Join krnlyng [0] (~liar@
19:33:17 Join lebellium [0] (
19:37:55 Join b0hoon [0] (
19:41:19 Quit Strife89 (Ping timeout: 265 seconds)
19:45:03b0hoonpamaury: i had a working slice of code for the lcd of View but i've lost it, i could try to recreate it maybe slowly... :)
19:45:26b0hooneverithing else is a black magic and it must be reversed
19:45:58b0hooni stuck on the ATA controller and gave up and then i've lost the code fot the LCD :S
19:52:52 Join amayer [0] (
19:52:54 Quit amayer (Read error: Connection reset by peer)
19:53:09pamauryb0hoon: that's a shame :-/
19:54:26pamauryI've currently reversing the bootloader
19:58:14b0hoonyeah i was too fast in reinstalling linux distro
19:59:09pamauryit is a bit of a pain that the BL is written in C++ but at least it's compiled with type information so it's makes the job much easier
20:00:02b0hoonbut first of all, original bootloader needs NVIDIA,0,0 magic in some place i don't exactly remember in order to boot our crafted mi4 image
20:02:07 Join Rower [0] (
20:04:15b0hoonactually it's a check before writing it to flash through the recovery mode
20:04:42b0hoonbut i don't remember details now
20:04:46pamaurysomething is unclear to me: are the registers shared by the cop and cpu or are some duplicated ?
20:05:04pamauryfor example the MMAP_*
20:06:53b0hoondon't know really... there is no any documentation about SoC
20:07:36b0hooni know for sure that the 8-bit ports are to the T letter :D
20:08:17b0hoonon the first bit of port T is the state of the power button IIRC
20:08:27b0hoonor maybe HOLD
20:09:24 Join amayer [0] (
20:09:33 Quit amayer (Read error: Connection reset by peer)
20:15:27 Join b0hoon_ [0] (
20:17:49 Quit b0hoon (Ping timeout: 255 seconds)
20:25:58 Join amayer [0] (
20:28:19 Quit amayer (Read error: Connection reset by peer)
20:29:51 Join amayer [0] (
20:36:14 Quit amayer (Read error: Connection reset by peer)
20:37:39 Join amayer [0] (
20:37:44 Quit krnlyng (Remote host closed the connection)
20:39:29 Quit mc2739 (Ping timeout: 264 seconds)
20:39:49***Saving seen data "./dancer.seen"
20:41:02 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
20:42:36 Quit amayer (Ping timeout: 240 seconds)
20:55:37 Join amayer [0] (
20:56:52 Quit Rower (Read error: Connection reset by peer)
20:57:50fs-bluebotBuild Server message: New build round started. Revision f8d367b, 253 builds, 32 clients.
21:02:35fs-bluebotBuild Server message: Build round completed after 285 seconds.
21:03:17 Quit ygrek (Ping timeout: 265 seconds)
21:03:55b0hoon_:O wtf
21:07:23gevaerts[Saint]: ping
21:07:53gevaerts[Saint]: your build client is broken
21:09:54b0hoon_on storebror-daniel it looks worse
21:10:12gevaertsYes, but bagder's not in here. I pinged him in #rockbox-community
21:10:48*gevaerts blocks both of them for now
21:26:38fs-bluebotBuild Server message: New build round started. Revision ee1a1ce, 253 builds, 31 clients.
21:29:48 Join newbie|3 [0] (
21:33:28fs-bluebotBuild Server message: Build round completed after 410 seconds.
21:33:47 Quit RiD (Ping timeout: 255 seconds)
21:33:56 Quit |akaWolf| (Ping timeout: 265 seconds)
21:34:13b0hoon_now it's better
21:38:02 Join habys [0] (
21:38:07newbie|3Apparently I tried to often to "recall" my account password and now my account is said to be banned because my nickname sounds like spam...?! Very strange. Anyways, I'd like to get my account unbanned, please. Thanks!
21:38:54newbie|3Uuups, KVirc did change my nickname, it's supposed to be DTSyX
21:39:04 Join |akaWolf| [0] (~akaWolf@
21:39:25gevaertsWhich account?
21:40:27newbie|3same name: DTSyX
21:41:04gevaertsAh, found it...
21:41:24 Join tertu [0] (
21:41:28gevaertsSomeone banned forums*@*
21:41:34newbie|3rockbox. isn't this the #rockbox irc chat? That would explain why my nick isn't correct either...
21:41:36*gevaerts removes that
21:41:49gevaertsnewbie|3: it is, but we have accounts in more places :)
21:41:59gevaertsAnyway, working on it
21:42:20newbie|3Ah, I see. I never used irc before... never had to :-)
21:42:28newbie|3Ok, thanks a lot!
21:42:59gevaertsUnfortunately handling bans on the forums is *slow*, so I'll need a few minutes...
21:44:28newbie|3No worries, I'm not in a hurry. I was just changing my email address and apparently after the change I was mistyping the right password so I tried a lot of wrong ones and got banned...
21:45:17gevaertsIt's not that. We've had a spam wave a few weeks ago, and some of the resulting bans were overly broad
21:46:22newbie|3Ok. That's a unpleasant coincidence. :-(
21:46:53*gevaerts nods
21:48:26*gevaerts taps his foot impatiently waiting for the forums to respons properly
21:50:28 Join RiD [0] (Ghost@
21:50:29newbie|3Well, I see the problem now: I changed my email address to one that contains "forum". That would explain it. With all the data leaks in the last few month I wanted to have different email addresses for different purposes to be maybe able to see were (future) leaks are coming from.
21:51:07gevaertsAh, yes. That explains the entire thing :)
21:51:24gevaertsscorche|sh: we really need to fix the forums in some way...
21:51:36saratogai think i banned spammers from doing [randomuser]
21:51:44gevaertsnewbie|3: can you try again?
21:51:53gevaertssaratoga: this one was on forums*@*
21:51:53saratogathere was some bot net that liked doing that
21:52:07gevaerts14 hits, I imagine most of those being from newbie|3
21:52:24gevaertsAnd the spam wave is over now anyway
21:52:40amiconnpamaury: pong
21:53:08*amiconn also just updated the android toolchain for his build client
21:54:14newbie|3It works now, I got in. Thank you very much! 14x, that seems about right... If the first try is a typo you try every other one exept the right one again... classic...
21:55:04newbie|3So, I can keep the "forum*@*" mail address for my account?
21:56:47newbie|3Great. Thanks!
21:57:10gevaertsYou're welcome, and sorry about the trouble
21:58:30newbie|3That's ok. That way I finally had to use and get to know irc... :-)
22:00:22 Quit Zambezi (Quit: Suspended: Expired)
22:02:06 Quit newbie|3 ()
22:02:16 Join DTSyX [0] (
22:02:33 Quit RiD (Read error: Connection reset by peer)
22:04:25 Quit DTSyX (Client Quit)
22:05:08 Join DTSyX [0] (
22:06:56 Join RiD [0] (Ghost@
22:07:12 Join rela [0] (~x@pdpc/supporter/active/rela)
22:12:25DTSyXThanks again and good bye.
22:12:42 Quit DTSyX ()
22:14:15 Quit the-kyle (Ping timeout: 240 seconds)
22:23:39 Join ZincAlloy [0] (
22:23:46scorche|shgevaerts: ah - i thought we werent having much spam lately
22:24:00gevaertsscorche|sh: no, that part is fine
22:24:14gevaertsBut doing anything with the ban list is still horrendous
22:24:57scorche|shgevaerts: either way, there shouldnt be much banning necessary - for user bans, move them to the banned membergroup...for IP and email bans, you do need to ban, but those should be pretty few and far between
22:25:32gevaertsThis one was about removing a ban :)
22:25:35ParkerRrasher, thanks for the Android builds. Kinda fun to use on my Nexus 7 :D
22:25:42scorche|shgevaerts: ah - there is that...
22:26:06lebelliumwhaouh! cabbiev2 for YH-820! Never late than never :)
22:26:25ParkerRNever late than never haha
22:29:19rasherParkerR: My contribution is minimal - but thanks still
22:32:08ParkerRrasher, Well considering I'm too lazy to set a toolchain up, it helps :)
22:39:52***Saving seen data "./dancer.seen"
22:45:40 Quit ender` (Quit: The likelihood of Perl being involved in a system is directly proportional to the length of time the system has been in maintenance.)
22:57:26 Quit rela (Read error: Connection reset by peer)
22:59:26ParkerRIs it possible to save a recording to a folder other than root? I've tried a few keypresses while hoverded on recording name but can't seem to get a kyeboard up to type anything
23:00:19ParkerROh theres a set as recording directory option for hovering on a directory
23:01:28ParkerRCtrl + F'ing the manual sure is nice
23:19:08 Join the-kyle [0] (
23:24:59 Quit amayer (Quit: Leaving)
23:39:19 Quit kuldeepdhaka (Quit: alha aphis)
23:40:36 Quit lebellium (Quit: ChatZilla [Firefox 29.0/20140324101726])
23:45:28fs-bluebotBuild Server message: New build round started. Revision 0eae33c, 253 builds, 30 clients.
23:47:53 Quit pamaury (Ping timeout: 264 seconds)
23:52:36fs-bluebotBuild Server message: Build round completed after 428 seconds.
23:53:48 Part b0hoon_ ("GTG... Bye.")

Previous day | Next day