--- Log for 22.08.114 Server: verne.freenode.net Channel: #rockbox --- Nick: logbot Version: Dancer V4.16 Started: 2 days and 23 hours ago 00.01.15 # dfkt: that bbe-ish DSP wouldn't help the group delay. because an extra dsp means more delay. 00.01.35 # besides I've make one before. https://github.com/papaloco1234/Rockbox-RCC/blob/notRCC/0815/lib/rbcodec/dsp/bbe.c 00.01.49 # wow... the doom plugin is *dirty* 00.02.01 # Everything is indented by 3 spaces... 00.02.37 # gevaerts, would you mind if I just ran the indentation script on the DOOM plugin, not the whole codebase? 00.03.11 Quit pamaury (Remote host closed the connection) 00.03.18 # ploco, did you reverse the original BBE, or how did you get the numbers? 00.04.02 # I get those group delays figures from WIki.LOL 00.06.06 # the whole idea is to make high band faster than mid band, follow by low. and the result is a *much sharper sound 00.06.38 # ploco, do you know much about the audio playback system? 00.07.16 # audio playback system? 00.08.06 # Like what is the pcmbuf format? 00.08.14 # i've used that CLAS/BBE VST thing in fb2k with good results. it certainly is more subtle than the BBE D82 VST plugin 00.10.12 # franklin: its all in the recodec/dsp 00.10.23 # rbcodec? 00.11.10 # dfkt: yes, my version would sounds like shit too if without the bishelf_coefs boost. 00.11.33 # <[Saint]> It sounds pretty out there, I know, so hold on to your hate...you might need to sit down for this one...the PCM buffer's format is...shockingly, PCM. 00.11.43 # <[Saint]> *hold on to your hat 00.13.01 # <[Saint]> (please don't ask what PCM is...) 00.13.24 *** Saving seen data "./dancer.seen" 00.13.48 # pcm is the only output format Rockbox supports. unlike DSD 00.19.37 # btw, the exciters extract some frequency band out from sample, phase delay then add it back to the original. 00.21.23 # some claim this will repair the harmonic to its best. but more like introduce "missing fundamental" virtual base band. 00.35.40 Quit sakax (Remote host closed the connection) 00.40.52 Quit bertrik (Remote host closed the connection) 00.43.42 Quit kuldeepdhaka (Quit: going to sleep before anyone else wakes up!) 00.46.55 Nick DormantBrain is now known as SuperBrainAK (~andy@74.112.200.73) 00.49.58 Quit ploco (Quit: Page closed) 01.06.38 Nick SuperBrainAK is now known as DormantBrain (~andy@74.112.200.73) 01.10.56 # [Saint], so the void* pointer returned by pcm_request_buffer is... what? 01.11.00 Quit ZincAlloy (Quit: Leaving.) 01.12.55 # [Saint], so it's a void pointer to... ints? unsigned ints? shorts? ushorts? chars? uchars? 01.19.18 # [Saint], so how do I know the samplerate/bit depth? 01.19.57 # never mind... pcm_get_frequency() :) 01.32.53 # * [Saint] stabs himself in the face for attempting extended locale support again 01.33.48 # Stabbing oneself in the face is a bad idea. 01.39.57 # TheSeven, can you walk me through how PCM playback works? From what I can gather from the source, INT_DMAC0C0 copies the Rockbox PCM buffer to the DMA buffer used by the audio chip 01.40.01 # right? 01.53.09 # TheSeven: ping 01.56.18 Quit alexbobp (Ping timeout: 250 seconds) 01.56.28 Join alexbobp [0] (~alex@capitalthree.pwnz.org) 02.13.26 *** Saving seen data "./dancer.seen" 02.15.47 Quit Jinx (Quit: reboot) 02.37.29 Quit krabador (Quit: Take the time.) 02.58.43 Join saratoga_ [0] (123e1c18@gateway/web/freenode/ip.18.62.28.24) 02.59.13 # ploco: what is TTS? 03.00.05 Quit AlexP (Read error: Connection reset by peer) 03.01.21 # Text to speech? 03.01.37 # Has anyone considered adding a tic-tac-toe game yet? 03.01.38 # its some kind of DSP effect 03.01.50 # ah 03.02.13 # text to speech? 03.02.45 # RiD, just said that ;) 03.02.58 # ._. 03.03.11 # i probably need glasses 03.03.36 # I mean, seriously... there's DOOM and no tic-tac-toe? 03.03.39 # (and chess) 03.04.54 # I know of the CHIP8 tic-tac-toe game... but it is terrible! 03.04.58 # No AI... 03.04.59 # who the hell wants tic tac toe? Doom is the sh_t! 03.05.08 Join ploco [0] (dce9b7f9@gateway/web/freenode/ip.220.233.183.249) 03.05.10 # what aboout 2048? 03.05.55 # TTS, temporary threshold shift, should have name the dsp anti-TTS... 03.06.09 # y'all experts, what would be needed to somehow port pictureflow to another rockbox-capable device? 03.06.18 # actually.. rockbox as an app. Nokia N900. 03.06.35 # Does android even support plugins? 03.07.04 # If not, then it won't be easy 03.07.10 # saratoga, do you know? 03.07.22 # no idea... but not running android, maemo 03.07.26 # i don't think there are plugins on android 03.07.46 # ploco: i'm not familiar with that, but basically its some EQ settings? 03.08.02 # http://www.rockbox.org/wiki/MaemoPort does not say anytthing about plugins... must dig deeper 03.08.12 # <[Saint]> Bah... 03.08.31 # I'm guessing no 03.08.41 # <[Saint]> I should've left it when I decided I wasn't going to do RTL locale support. 03.08.46 # <[Saint]> This is driving me nuts. 03.08.48 # <[Saint]> :-/ 03.08.54 # then stop :) 03.09.02 # ploco: were you going to put that code on our gerrit? i think it looks quite good 03.09.20 # <[Saint]> Indeed. 03.09.44 # <[Saint]> A BBE clone has been requested a few times. 03.09.54 # <[Saint]> ...what would we call it, though? 03.10.02 # <[Saint]> BBE is trademarked, surely. 03.10.04 # although the BBE people might prefer if we didn't use their name in the code 03.10.13 # <[Saint]> heh - snap. 03.10.15 # what is BBE? 03.10.21 # <[Saint]> ask google 03.10.39 # "bad boy entertainment"? 03.10.49 # i'm not even sure if its the same thing, i've seen some of the bass boost effects in various OFs, and they're usually a combination of filtering and other stuff 03.11.01 # although the effect is probably quite similar 03.11.13 # Bachelor of Built Environment? 03.11.30 # Plugin returned error 03.11.33 # hmm ;\ 03.11.42 # So it does build plugins? 03.11.42 # i'm guessing that's a no 03.11.58 # If the plugin returns an error code, then it must build plugins :) 03.12.11 # a good thing? 03.12.18 # I don't know. its just a group delay correction with a 5db bass boost and 3db treble gain 03.12.29 # I mean the bbe 03.12.41 # What does it stand for? 03.12.47 # franklin would the rockbox credits.rock be considered a plugin? 03.12.52 # because that works, lol 03.12.54 # RiD: yes 03.13.04 # If it ends with .rock, it's a plugin 03.13.12 # ploco, is http://www.bbesonicmax.com/ the URL? 03.14.14 # franklin is pictureflow different, depending on the device? 03.14.21 # what do you mean? 03.14.30 # yes, that's the bbe people talk about. 03.15.00 # ah 03.15.02 # i'm a noob at this, but i guess pictureflow is also a .rock, and if so, do they differ from device, or is it the same... uh, .rock for all? 03.15.32 # RiD, .rock is an executable format, so it contains instructions specific to the device's hardware (CPU/RAM, etc.) 03.15.45 # So no, it differs from device to device 03.15.45 # oops 03.16.32 # well, thanks for clearing that up :) 03.16.33 # So you're making a clone of SonicMax Pro? 03.16.42 # RiD, the same goes for EXE's 03.16.46 # And any executable format 03.16.57 # guess it's not worth the hassle, and i probably can't do anything with my current knowledge about all this 03.17.12 # (i.e., a .elf for an ARM platform won't work on a, say, x86 platform) 03.17.34 # RiD, what does it say before the "Plugin Returned Error" message? 03.17.49 # You probably just need to initialize database 03.17.50 # on pictureflow? Nothing, because it does not exist 03.18.02 # it does not "ship" with it 03.18.22 # What does not "ship" with it? 03.18.57 # pictureflow. It's not in the rocks folder and subfolders (dapps, demos, games, viewers) 03.19.34 # the only ones it has are viewers: credits, lua, mikmod, properties, search, sort, theme_remove, vbrfix. And not all of them work 03.19.45 # sorry, *apps not dapps 03.20.12 # What's under demos? 03.21.00 # nothing at all 03.21.13 # games? 03.21.15 # nor apps & games 03.21.22 # <[Saint]> OK. Awesome. Basic RTL locale support completed. 03.21.35 # congrats... no stabbing yourself in the face? :) 03.21.39 # <[Saint]> Uuuuugh. Fuck yeah. Finally. 03.21.55 # <[Saint]> And no - no face stabbing. 03.22.05 # <[Saint]> But, it was touch and go there for a while... 03.22.42 # RiD, what does the Maemo port even look like? I have no experience with any of the RaaA platforms 03.22.53 # <[Saint]> Code looks hideous though. 03.23.04 # franklin, look like, as in? 03.23.06 # <[Saint]> I bet if I took all the comments out it might be enough to make people weep. 03.23.24 # Not the code, just what are you running it on? 03.23.48 # Like a screenshot? 03.24.08 # ah you want a screen? i can do that, *derp* 03.24.38 # http://my-maemo.com/grafika/rockbox.jpg - not too different from that 03.24.56 # Like what platform are you on (hardware-wise)? 03.25.09 Quit xaioscn (Remote host closed the connection) 03.25.19 # <[Saint]> http://pastebin.com/N7C2Xz6j 03.25.21 # saratoga_: I am trying to think of a simple way to describe the TTS......by significantly reduce the frequencies that may triggers the Auditory fatigue, the human hearing threshold should stay in a sweet spot for a long time.... 03.25.25 # <[Saint]> 368 lines of pure terror 03.25.32 # ARM, TI OMAP 3400 (600Mhz Cortex A8, has a PowerVR SGX530), 256MB RAM 03.25.42 # ah nice 03.25.59 # 3400? oops. 3430* 03.25.59 # But as for getting pictureflow to work, I can't do much 03.26.14 # ploco: ok thats what i understood from wikipedia, but does it sound good to listen to music like that? 03.26.22 # nah i was just wondering, not really something i'd want. (Although it'd probably look nice) 03.26.32 # ploco, where's the code? I'd love to try it 03.26.39 # <[Saint]> franklin: runs on N900 03.26.43 # franklin: https://github.com/papaloco1234/Rockbox-RCC/commit/9c575f56301aa6ef4ce57b3b0c2f0cae5bbb2b3b 03.26.55 # <[Saint]> ...and that other obscure maemo device no one cares about 03.27.00 # N8xx haha 03.27.04 # ooh... drool 03.27.09 # saratoga_: yes, it sounds awesome once hearing get use to the new low/high ratio 03.27.17 # <[Saint]> N900 /was/ a nice phone. 03.27.21 # <[Saint]> ANd it still is really. 03.27.26 # <[Saint]> Its just bloody dated. 03.27.31 # probably not good as a phone 03.27.40 # but uh, i can live with it for a few more years if it doesn't die on me 03.27.44 # <[Saint]> It was. 03.27.49 # ploco, so any way that you might be able to get it on Gerrit? 03.28.05 # i'm surprised there's still some support for it, so kudos for that 03.28.10 # it's been 5 years 03.28.13 # (4 for me) 03.28.38 # <[Saint]> There's even a whole other spinoff using the same board. 03.28.41 # <[Saint]> Neo900 03.28.48 # ya 03.28.54 # <[Saint]> foolishly expensive though. 03.28.55 # franklin: if you can provide two better names for these two dsps 03.28.56 # bit out of my budget, but i hope the best for it 03.29.10 # I have no idea what they do :) 03.29.38 # So I couldn't name them 03.29.47 # another crazy question. does Rockbox as an app have some sort of commandline? so that you can tell it to play this file, bla bla? 03.30.02 # <[Saint]> Nope. 03.30.20 # I was gonna port bash... but then I realized I would need so much other stuff 03.30.28 # aw 03.30.32 # can just get my source from here and build it yourself https://github.com/papaloco1234/Rockbox-RCC/ 03.30.35 # <[Saint]> But, if you can access a commandline on the device natively, and have write access, you could compile Warble for the host 03.30.55 # the branch is notRCC/0815 03.31.13 # I don't even know what warble is, so yeah :\ 03.31.30 # ploco, it'll be done in a few minutes... 03.31.32 # <[Saint]> its a standalone Rockbox playback library 03.32.07 # <[Saint]> all our codecs, basic playlist support, commandline based sekking via offset 03.32.12 # RiD, if you want, I could help you write a simple shell... 03.32.25 # builtin ls/cd/cat/etc. 03.32.38 # It wouldn't be that hard... 03.33.08 # (except for the fact that RB only supports absolute paths) 03.33.26 # BTW, is there any chance that relative paths could be supported? 03.36.44 # ploco, I think it'd be a great idea to get the code on Gerrit. that way, more people could test it 03.37.12 # eh, don't know... Would probably be cool in the long term (to have a native gui controlling rockbox), but i'm very lazy, and zero experience on coding an UI other than... a rockbox skin 03.38.03 # * franklin is really pondering the cause for the lack of a tic-tac-toe game 03.40.16 # franklin, i will try to put them up later tonight. 03.41.27 # <[Saint]> Call them "NotBBE" and "NotTTS" ;) 03.42.12 # What do they do? What are they, even?! 03.42.55 # <[Saint]> "enhancement" (debatable) DSP effects 03.43.05 # ah 03.44.15 # [Saint], do you know much about the wiki? 03.44.25 # <[Saint]> No. 03.44.45 # I just love when you say "No." 03.44.50 # So... to the point 03.45.19 # Just like "Is there a way to modify the theme "... "Yes." 03.45.30 Join swagglypuff [0] (~Kazo@24.224.253.233) 03.46.06 # <[Saint]> Well, I didn't lie. He didn't ask /how/ to do it. ;) 03.46.11 # haha 03.46.27 # Is it possible to ? 03.47.52 # Yes. 03.49.23 Join Water255 [0] (~chatzilla@108.87.16.229) 03.50.44 # Sansa Fuze+... i notice USB connect now defaults to RB instead of the OF. Just curious what the breakthrough was. Did we finally get a datasheet? 03.51.11 # Oh and RB transfers files fast (like the OF) too. 03.53.20 # <[Saint]> No datasheet, no. At least, not that I'm aware of. 03.53.42 # <[Saint]> pamaury is just a master of reverse engineering. 03.54.38 # As are most of the RB porters 03.55.27 # <[Saint]> pamaury and wodz take it to a whole other level. 03.55.53 # Like how? 03.56.19 # i think the fuze+ had a well documented SOC 03.56.49 # <[Saint]> Just in terms of dedication and knowledge. Well, that's my opinion, anyway. 03.56.53 # What I'm curious about is... how do you even get code to run? 03.57.10 # Find a buffer overflow somewhere... how? 03.57.26 # ploco: "perceptual bass enhancement" and "Fatigue Reduction"? 03.58.12 # <[Saint]> I'm quite interested in your surround DSP also. 03.58.14 # franklin: the new ports thread, wiki and IRC logs usually record how code execution was achieved for most devices 03.58.29 # <[Saint]> but it looks like quite some effort to cut that out. 03.59.37 # ho, cool name. 04.00.20 # <[Saint]> man - we /really/ need to walk over your git repo 04.00.38 # <[Saint]> there's a great number of trivial commits worth stealing. 04.00.41 # saint, haas surround? that is a toy make you WOW for about.....2 mins? 04.01.27 # haha 04.01.30 # <[Saint]> Indeed. We can get an approximate effect (ab)using stereo width too, I guess. 04.01.44 # a lot of them are experimental DSP only for fun. 04.02.00 # <[Saint]> That's true of pretty much all DSPs really. ;) 04.02.49 # I made a hrtf with no delay dsp before but 99% users doesn't feel a thing. 04.04.16 # * [Saint] begins making a list of commits to burgle 04.04.34 # "burgle" 04.05.01 # <[Saint]> hahahahahaha 04.05.18 # lololol 04.05.31 # * [Saint] sees that ploco hated the list line separator so much that merely disabling it wasn't sufficient 04.05.51 # <[Saint]> (a sentiment I personally share) 04.06.18 # <[Saint]> errr, PurlingNayuki, rather 04.06.39 # what commit was that? 04.07.01 Quit Water255 (Quit: ChatZilla 0.9.90.1 [Firefox 32.0/20140811180644]) 04.07.06 # that's because I "Drawing the lists using a skin" for my theme. list line separator doesn't matter at all 04.07.31 # <[Saint]> franklin: c4d6c18 04.07.54 # you can't search by sha1 on github!?!? 04.08.01 # <[Saint]> ploco: I just think it looks weird... 04.08.57 # ploco, I get an undefined reference to filter_bishelf_coefs in bbe_configure, dsp_bbp_precut, and strength_update 04.10.09 # make clean and make reconf 04.10.20 # ploco, this is a fresh clone 04.10.34 # so I don't think that would help 04.11.29 # Regarding G#910, I think the real bug is the order of the icons in the icon sets 04.11.32 # 3Gerrit review #910 at http://gerrit.rockbox.org/r/910 : 3Improved viewers.config by Franklin Wei 04.11.42 # dsp_bbe_precut is in that patch 04.11.59 Join Scr0mple [0] (~Simon@119.225.209.134) 04.12.24 # That's the function that's calling the undefined function 04.13.29 *** Saving seen data "./dancer.seen" 04.14.37 # #include "dsp_filter.h" < no, it's declared, not defined 04.15.17 Quit Scromple (Ping timeout: 245 seconds) 04.15.20 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn) 04.15.20 Quit amiconn (Disconnected by services) 04.15.22 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn) 04.15.43 Quit pixelma (Disconnected by services) 04.15.43 Join pixelma_ [0] (quassel@rockbox/staff/pixelma) 04.15.45 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma) 04.15.52 # <[Saint]> Hmmmm. "backdrop: -" in my theme .cfg isn't clearing the backdrop properly. 04.16.08 # <[Saint]> It keeps the old backdrop until a power cycle. 04.16.32 # ploco, it's declared in dsp_filter.c... maybe it's not being compiled in? 04.16.37 # ok... that function need this #ifdef HAVE_SW_TONE_CONTROLS 04.16.56 # so define HAVE_SW_TONE_CONTROLS 04.17.27 # ah 04.17.29 # where? 04.17.48 # just some random header? 04.18.10 Quit RiD (Quit: A good plan today is better than a perfect plan tomorrow.) 04.18.24 # <[Saint]> feel free to bang your head against the desk at any time ploco 04.18.30 # firmware/export/config than the one for your target 04.18.55 # haha 04.20.59 # [Saint]: I face worst in Chinese forum. paragraphs of meaningless words. ack like know everything inside RB but not even close. 04.21.29 # <[Saint]> :)) 04.22.31 # Ah great... now it defines __audiohw_setting_BASS twice 04.22.52 # some group of ppl ask a 31 band eq forever and non can actually use it right. god.. 04.23.32 # They want a 31-band equalizer?! 04.23.38 # Why not 32? :) 04.23.39 # franklin: maybe just try build a target that initially HAVE_SW_TONE_CONTROLS 04.24.01 # even 10 is more than enough. this is not graphic-eq 04.24.38 # fuze+, I guess 04.25.24 # Can the sim play audio? 04.26.13 # I've never tried it, actually... 04.26.27 # The most I use the sim for is to test out some plugins... :) 04.27.05 # mask #ifdef HAVE_SW_TONE_CONTROLS in dsp_filter.c and dsp_filter.h 04.27.29 # then this should be available even if not define HAVE_SW_TONE_CONTROLS 04.27.39 # *not defined 04.28.38 # <[Saint]> ploco: everyone uses it like a GEQ though. 04.29.06 # <[Saint]> The primary reason I added the extra bands is so people wouldn't need to piss around with the frequencies so much. 04.29.18 # <[Saint]> But people ended up just setting all of them and ignoring the q value. 04.29.27 # Made it 10 bands or 31 bands? 04.30.37 # <[Saint]> To be completely fair, at the time I didn't have a complete understanding of a full parametric EQ either. 04.31.34 # [Saint]: those are really hopeless people. nowdays I suggest them: go play Fruityloops and free vst before ask me any eq questions. 04.33.11 # ploco, it doesn't play 04.33.23 # the position remains at zero even with pause off 04.33.36 # (and yes, my volume is max) 04.34.16 # stop it.. do you have a android device? I got pre-compiled version for you 04.34.20 # no 04.34.26 # ipod classic 04.37.42 # btw, why is it a bad idea to use bitfields in rockbox? 04.38.03 # then shell wait till I put it on gerrit. 04.38.14 # <[Saint]> heh. 04.38.14 # <[Saint]> cabbiev2 bitmap assets: 3.2MB 04.38.14 # <[Saint]> touch_theme (cool name, huh? :)) bitmap assets: 700kB 04.38.46 # <[Saint]> it uses 1.3MB of fonts though. 04.38.57 # <[Saint]> but it still comes in a lot smaller. 04.39.30 # [Saint], let me guess... it's black and white... and it looks terrible... but it functions great? 04.40.34 # <[Saint]> Y'know? You make it really difficult to not mute you. 04.42.57 # * franklin is wondering how to implement PCM playback over the piezo 04.43.58 # Should I use a realtime thread... or an IRQ? 04.44.23 # Or just piggyback on the existing DMA transfer IRQ? 04.44.27 Quit derf (Ping timeout: 250 seconds) 04.45.50 # no... the DMA IRQ isn't called nearly fast enough... I guess a realtime thread would be best 04.46.13 # try to get it from one of the dsps. 04.46.30 # get what? 04.46.31 Join derf [0] (~derf@static-108-18-126-14.washdc.fios.verizon.net) 04.47.20 # plus, a realtime thread would simplify the code a lot 04.47.22 # <[Saint]> You're going to have a LOT of audio dropouts. 04.47.32 # the pcm samples, use dsp to copy them to your piezo 04.47.46 # ploco: now there's an idea 04.48.21 # though I don't know much about how it works? 04.48.23 # . 04.48.27 # s/?/./ 04.48.53 # <[Saint]> So, learn. Its all there. 04.49.24 # yep... that's what I'm doing :) 04.49.52 # ok. gtg. have fun coding :) 04.50.10 Quit ploco (Quit: Page closed) 04.50.13 # It's getting late here too 04.51.03 # ah so add source files under lib/rbcodec/SOURCES... 04.51.36 # Well... it can wait till tomorrow 04.51.39 # For now, g'night 04.51.42 Quit franklin (Quit: Leaving) 05.12.04 Quit steffengy (Disconnected by services) 05.12.04 Join steffengy1 [0] (~quassel@p5088FDD0.dip0.t-ipconnect.de) 05.16.19 Join Strife89 [0] (~Strife89@adsl-98-80-234-11.mcn.bellsouth.net) 05.19.12 # <[Saint]> So, IIUC, Android is fixed in head, but still needs HAVE_SIGALTSTACK_THREADS for ART support? 05.40.13 # [Saint], woo andorid fixed? :D 05.41.28 # [Saint], woohoo loaded! 05.41.32 # What was the issue? 05.43.18 # <[Saint]> A snafu with threading. 05.43.32 # https://github.com/Rockbox/rockbox/commit/a9713d89e76158c6b7be4d8873a921e30efe688f this? 05.43.56 # <[Saint]> yes 05.44.19 # <[Saint]> sorry, no. 05.44.22 # <[Saint]> http://git.rockbox.org/?p=rockbox.git;a=commit;h=5fb3702 05.44.43 # Ahh :) https://github.com/Rockbox/rockbox/commit/5fb370267f81a8429a76f0237767804b849d1f81 05.44.54 # I like the mirroring 05.44.58 # <[Saint]> that was one of the series of commits that uncovered this, though. 05.45.24 # <[Saint]> It didn't really /break/ the Android port. 05.45.41 # <[Saint]> Rather, it more or less pointed out that its a bit or a miracle it ever worked to begin with. 05.45.57 # <[Saint]> *bit of a 05.47.30 # Heh 05.48.14 # [Saint], so umm hows that theme coming along? :) 05.52.11 # <[Saint]> see for yourself: http://www.datafilehost.com/d/5c0868a1 05.52.29 Quit TheSeven (Ping timeout: 272 seconds) 05.52.35 # <[Saint]> It'll /try/ to scale to arbitrary resolutions, but some things are locked in place. 05.53.01 # <[Saint]> (and none of the sliders in the sound settings submenu are expected to actually function presently) 05.53.25 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven) 05.53.28 # :D 05.53.32 # <[Saint]> And there's some draw issues that I'm trying to find a nice way of fixing. 05.53.50 # So should it support 720x1280 fairly well? 05.54.06 # <[Saint]> it should be usable at least. 05.54.11 # <[Saint]> though it might look a bit odd. 05.56.30 # <[Saint]> I'm still planning on resolution specific versions. The self-healing scaling is mainly for the weird resolutions, like 480x854, or 720x1288, etc. 05.57.20 # <[Saint]> there a few px here and there won't make a huge visible difference, and its not worth creating s specfic theme for the sake of positioning everything _exactly_ right. 05.57.27 # <[Saint]> *where a 05.58.31 # <[Saint]> I have three things to finish now: 05.58.47 # <[Saint]> 1 - touchscreen settings callbacks, so that the settings sliders actually work 05.59.02 # Ummm 05.59.05 # Its really broken 05.59.06 # <[Saint]> 2 - fixing the draw artefacts 05.59.10 # Screenshot incoming 05.59.19 # <[Saint]> 3 - resolution specific versions 06.00.24 # [Saint], https://mediacru.sh/b858a3a005eb 06.01.05 # Or is it kinda mixing my old theme and the new? 06.01.07 # <[Saint]> ...what the 06.01.15 # I selected it witht he first menu in the theme settings 06.01.27 # "browse themes" 06.01.37 # Ill trya fresh folder 06.01.45 # <[Saint]> that's definitely got some leftovers, indeed. 06.01.56 # <[Saint]> lots of assets there my theme doesn't employ. 06.04.15 # <[Saint]> just caught one slight issue with the scrubber's touch area. 06.04.27 # <[Saint]> nothing that would cause /that/, though. 06.05.33 # [Saint], https://mediacru.sh/e32a28018c65 06.05.37 # Brand new folder 06.06.16 # "Browse theme files" selected touch 06.06.22 # <[Saint]> Yeah, that looks about what I would've expected. 06.06.28 # Ahh ok 06.06.36 # Works pretty ok :D 06.06.56 # Nice just having cntrol buttons instead of tapping in the middle of the screen 06.07.33 # [Saint], thanks 06.07.45 # <[Saint]> its only supposed to have to scale a couple of px in either direction - not several hundred. But it should be largely usable. 06.09.28 # [Saint], Android seems like the lonely target but you still have some love for it :) 06.10.05 # <[Saint]> Yeah. Its a bit of a niche. 06.10.41 # What other touchscreen targets are there? 06.11.46 # <[Saint]> maemo, SDL app, Open Pandora, Cowon D2 06.12.01 # Ahh nice 06.12.07 # <[Saint]> probably a few I'm missing. 06.13.30 *** Saving seen data "./dancer.seen" 06.13.46 # Which reminds me I was going to build the SDL to mess with 06.14.00 # [Saint], what res does that theme target atm? 06.14.07 # Best I mean 06.14.13 # <[Saint]> 480x800 06.16.30 # <[Saint]> It needs a name. 06.16.44 # <[Saint]> And NOT Rock* 06.16.48 # <[Saint]> or *box. 06.16.51 # <[Saint]> ;) 06.17.04 # [Saint], RockTouch 06.17.06 # Tock 06.17.39 # Toucbox 06.17.42 # :P 06.24.15 # [Saint], lol https://mediacru.sh/Muog9hyVbhv0.png 06.24.30 Nick DormantBrain is now known as SuperBrainAK (~andy@74.112.200.73) 06.24.36 # I call it art! 06.25.28 # <[Saint]> what the shit? 06.25.33 # I dunno 06.25.37 # Built 480x800 06.25.42 # <[Saint]> HALF NUKED CABBIE LEFTOVER BACKDROP? 06.25.46 # prefix is ~/bin/rockox 06.25.49 # Loads properly 06.25.50 # <[Saint]> sorry caps. 06.26.01 # Possibly 06.26.11 # I did the same "Browse theme" option 06.26.16 # Aha 06.26.27 # [Saint], it "fixed" with albuma art 06.26.36 # oh ol nvm 06.26.43 # album art just covered the glitches 06.26.47 # :3 06.28.02 # [Saint], It "may" be something screwy with my nvidia card. it does the same thing inbetween monitors... https://mediacru.sh/nTELc1jDTD21.png 06.28.17 # <[Saint]> [14:15:53] <[Saint]> Hmmmm. "backdrop: -" in my theme .cfg isn't clearing the backdrop properly. 06.28.17 # <[Saint]> [14:16:09] <[Saint]> It keeps the old backdrop until a power cycle. 06.28.31 # Hmm Ill try reatart 06.28.47 # Aha 06.28.52 # [Saint], lol that fixed it 06.28.54 # <[Saint]> Woo. :) 06.29.09 # <[Saint]> Yeah, seems like the backdrop isn't getting cleared properly. 06.29.23 # <[Saint]> exacerbated by the fact this theme doesn't use a backdrop itself. 06.29.51 # :D volume slider works 06.30.09 # <[Saint]> Everything should work apart from the sound settings sliders. 06.30.13 # lol I like how you got creative witht he menu 06.30.21 # Very nice 06.31.54 # [Saint], not too shabby :D https://mediacru.sh/BejnGKCfQXSe.png 06.33.31 # <[Saint]> If you happened to speak a RTL locale dialect, it'll reverse all the direction sensitive elements too. 06.33.41 # * [Saint] suspects that no one will use that feature, ever. 06.33.52 # Haha 06.34.20 # <[Saint]> I'm fairly confident that most people are quite used to UIs forcing English these days 06.35.58 # Rockbox SDl is actually kinda cool 06.36.01 # Nice for a simple player 06.36.11 # Nothing too fancy 06.36.41 # (I understdn its meanty for tetsing themes :P) 06.36.45 # *understand 06.36.56 # <[Saint]> yeah, I actually use it as my desktop player. 06.39.07 # [Saint], you should make it sdl2 for fullscreen detetection :P 06.39.17 # *detection 06.39.47 # [Saint], one thing I've noticed. The backdrop around the menu items doesn't turn grey until entering and exiting back out of a submenu item https://mediacru.sh/oWKVtMN1wIxJ 06.39.51 Quit Strife89 (Ping timeout: 245 seconds) 06.39.55 # btw that link is a video :) 06.40.27 # <[Saint]> yeah, that's: 06.40.28 # <[Saint]> [15:59:06] <[Saint]> 2 - fixing the draw artefacts 06.40.34 # Ahh 06.40.42 # <[Saint]> there's a couple of ways I can do it. 06.40.59 # <[Saint]> Not sure what the cleanest way is though. 06.43.48 Nick SuperBrainAK is now known as DormantBrain (~andy@74.112.200.73) 06.54.26 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS) 07.13.27 Join kugel [0] (~kugel@rockbox/developer/kugel) 07.38.02 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka) 07.39.25 Join JdGordon_ [0] (~jonno@rockbox/developer/JdGordon) 07.40.41 Quit JdGordon (Ping timeout: 260 seconds) 07.42.56 Quit swagglypuff (Ping timeout: 246 seconds) 07.45.23 Join swagglypuff [0] (~Kazo@24.224.253.233) 08.11.09 # <[Saint]> Nope. Bums. 08.11.16 # <[Saint]> Still borked in ART. 08.11.23 Join LinusN [0] (~linus@giant.haxx.se) 08.13.07 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon) 08.13.16 # <[Saint]> *PANIC* 08.13.16 # <[Saint]> Stkov power 08.13.16 # <[Saint]> pc:47ac0fb0 sp:47cc9240 08.13.16 DBUG Enqueued KICK [Saint] 08.13.16 # <[Saint]> bt end 08.13.33 *** Saving seen data "./dancer.seen" 08.13.40 Quit JdGordon_ (Ping timeout: 260 seconds) 08.15.07 # <[Saint]> here's the mapfile: http://pastebin.com/raw.php?i=3uNShrjR 08.15.58 # <[Saint]> jhMikeS, kugel (logs) ^ 08.16.18 # with the gerrit patch? 08.16.29 # <[Saint]> that's git head with HAVE_SIGALTSTACK_THREADS using the ART runtime on Android 08.18.28 # <[Saint]> gah - such a PITA changing runtimes and/or clearing the ART runtime cache. 08.19.04 # <[Saint]> even on a test device with a very minimal ROM and only Rockbox installed it still takes ages. 08.19.24 Nick DormantBrain is now known as SuperBrainAK (~andy@74.112.200.73) 08.19.57 # might just need a bigger default stack size 08.20.10 # <[Saint]> Hmmmm. 08.20.52 # <[Saint]> I'll give that a go later on this evening. Need to go do the grocery shopping for now. 08.20.56 # raise it up until it runs and see what the usages are; if too low lower it, if too high raise it some more 08.21.18 # <[Saint]> Just thought I'd let you guys know, as its not terribly often you can get a full panic address with a map file as well. 08.22.04 # <[Saint]> sorry, what? 08.22.22 # <[Saint]> I think I know what you mean but I'm having a hard time parsing that. 08.22.53 # I take it that it didn't start? 08.23.04 # <[Saint]> No, it did not. 08.23.12 # <[Saint]> Just an immediate panic. 08.23.22 # Just like, double DEFAULT_STACK_SIZE or something. Keep going till up until it boots 08.24.03 # Then check OS stacks, see if the use percentages are really low, still really high or about right and adjust till they seem reasonable 08.24.29 # <[Saint]> OK. I'll give that a go later on this evening. 08.26.13 # or just adjust the one thread (unfortunately there aren't target-specific extra bytes for particular stacks) 08.27.36 # the power stack looks very minimal (half the default except on JZ4732) 08.28.12 Join ender` [0] (krneki@foo.eternallybored.org) 08.28.38 # perhaps it calls into java for battery numbers 08.30.02 # perhaps more stack is needed with ART 08.30.05 # <[Saint]> perhaps interestingly, if you force stop it, and run it again, it _doesn't_ panic. 08.30.17 # <[Saint]> ...it just blackscreens, and locks up the device. ;) 08.31.11 # btw, why is AndroidPort not mentioned at http://www.rockbox.org/wiki/TargetStatus , is that intentional? 08.32.15 # [Saint]: locks the whole device? seem like fertile grounds for some kind of exploit 08.32.33 # <[Saint]> if only I wasn't already rooted. ;) 08.32.55 # are you running as root all the time. I don't. 08.33.06 # (not that I want to distract you from real development with silly suggestions, just curious: can valgrind on Android be useful for panics like that?) 08.33.13 Join ploco [0] (dce9b7f9@gateway/web/freenode/ip.220.233.183.249) 08.33.16 # <[Saint]> I'm rooted. But, its disabled 99% of the time. 08.34.13 # unless needed, USB debugging: off, root: no way! 08.35.01 # saint you test it without gerrit patch right? 08.35.09 # <[Saint]> kinda offtopic - but with a vaguely current version of Android, and the lockscreen you should definitely have, USB debugging on isn't any issue at all. 08.35.25 # <[Saint]> The device needs to be unlocked and the debug instance approved. 08.35.47 # <[Saint]> ploco: yes 08.35.55 Join petur [0] (~petur@rockbox/developer/petur) 08.37.40 # apps/hosted/android/notification.c < what does the power thread do in apps/hosted/android/notification.c? 08.40.01 # RockboxService_class/instance need not be global for the power thread 08.40.35 # the battery status are notifications 08.40.48 # these are local to the pthread that hosts all rockbox threads, rockbox threads can use them even if local 08.42.11 Quit Scr0mple (Read error: Connection reset by peer) 08.43.10 # ploco: not it git master 08.45.09 Join Scromple [0] (~Simon@119.225.209.134) 08.45.10 # hmm, guess I mess up two different branches 08.45.16 # <[Saint]> I assume that's from following the map file? 08.51.28 Join Zagor [0] (~bjst@80.239.169.202) 08.51.28 Quit Zagor (Changing host) 08.51.28 Join Zagor [242] (~bjst@rockbox/developer/Zagor) 08.54.16 Join Scr0mple [0] (~Simon@161.43.73.67) 08.57.21 Quit ploco (Quit: Page closed) 08.57.31 Quit Scromple (Ping timeout: 250 seconds) 09.11.58 Quit lleeloo (Ping timeout: 272 seconds) 09.41.11 Quit kuldeepdhaka (Ping timeout: 250 seconds) 10.00.36 Join swilde [0] (~wilde@euarne.ipv6.intevation.de) 10.00.50 Part swilde 10.02.42 Join einhirn [0] (~Miranda@p3E9E60EF.dip0.t-ipconnect.de) 10.11.50 Quit einhirn (Ping timeout: 255 seconds) 10.13.36 *** Saving seen data "./dancer.seen" 10.15.38 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 10.44.07 Join AlexP [0] (~alex@rockbox/staff/AlexP) 10.46.05 Join dys [0] (~user@2a01:1e8:e100:8296:21a:4dff:fe4e:273a) 11.05.20 Join babylonlurker [0] (~quassel@veda.xs4all.nl) 11.08.27 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de) 11.15.00 Nick SuperBrainAK is now known as DormantBrain (~andy@74.112.200.73) 11.59.03 Join ploco [0] (dce9b7f9@gateway/web/freenode/ip.220.233.183.249) 12.01.31 # sorry, my bad. power thread panic got something to do with firmware/target/hosted/android/powermgmt-android.c 12.02.39 # _battery_level use a instance created from RockboxService_instance 12.05.43 Quit ploco (Quit: Page closed) 12.13.41 *** Saving seen data "./dancer.seen" 12.15.39 Join krabador [0] (~krabador_@unaffiliated/krabador) 13.41.25 # ploco (logs): RockboxService_class/instance need not be global for the power thread (i already mentioned that) 13.42.16 Join ZincAlloy [0] (~Adium@pD9EEAEE5.dip0.t-ipconnect.de) 14.02.41 Quit einhirn (Ping timeout: 246 seconds) 14.11.46 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka) 14.11.47 Join ploco [0] (dce9b7f9@gateway/web/freenode/ip.220.233.183.249) 14.13.26 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs) 14.13.43 *** Saving seen data "./dancer.seen" 14.15.30 # kugel: once again, my assumption was only based on ADB log of where the crash point is. 14.19.10 Quit kuldeepdhaka (Max SendQ exceeded) 14.20.49 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka) 14.31.22 Quit pamaury (Quit: No Ping reply in 180 seconds.) 14.32.30 Join pamaury [0] (~quassel@rockbox/developer/pamaury) 14.34.10 Quit kuldeepdhaka (Max SendQ exceeded) 14.35.55 Join kuldeepdhaka [0] (~kuldeepdh@unaffiliated/kuldeepdhaka) 14.41.22 Quit krabador (Quit: Take the time.) 14.44.00 Join krabador [0] (~krabador_@unaffiliated/krabador) 14.52.50 Quit krabador (Quit: Take the time.) 14.52.55 Part ZincAlloy 14.56.35 Join RiD [0] (~RiD@bl22-12-42.dsl.telepac.pt) 14.58.08 Join krabador [0] (~krabador_@unaffiliated/krabador) 15.03.05 Join amayer [0] (~amayer@mail.weberadvertising.com) 15.37.07 Quit GeekShadow (Ping timeout: 240 seconds) 15.37.19 Join GeekShadow [0] (~antoine@nzf.turmel.info) 15.37.19 Quit GeekShadow (Changing host) 15.37.19 Join GeekShadow [0] (~antoine@reactos/tester/GeekShadow) 15.39.51 Quit kugel (Ping timeout: 245 seconds) 16.01.59 Quit Zagor (Quit: Clint excited) 16.02.32 # g#922 16.02.35 # 3Gerrit review #922 at http://gerrit.rockbox.org/r/922 : 3two new DSPs by Chiwen Chang 16.03.25 # now bbe is now pbe, tts become afr. 16.13.45 *** Saving seen data "./dancer.seen" 16.15.12 Quit ploco (Quit: Page closed) 16.23.11 Part LinusN 16.25.52 Quit petur (Quit: *plop*) 16.38.11 Quit cmhobbs (Ping timeout: 245 seconds) 16.45.07 Join franklin [0] (~franklin@unaffiliated/franklin) 16.45.34 Join suYin [0] (suYin@server1.bouncer4you.de) 16.46.32 # franklin: responding to logs, yes the SoC of the Fuze+ is well documented but 1) it doesn't help for the non-SoC part of the device 2) the documentation doesn't describe how boot works (well it didn't when we started the port) which made code execution quite hard 16.47.02 # ah thank you 16.48.50 # also the documentation only apply to the STMP3780/imx233 and not exactly to the STMP3700, which I had to reverse engineer a lot (I finally got the datasheet but much later) 16.49.20 # wow... were you the only one working on it? 16.51.20 # yeah mostly, I received some help here and there but it's hard to split the work 16.51.30 # well, thank you 16.58.32 Join Rower [0] (husvagn@h176n2-aeg-a11.ias.bredband.telia.com) 17.00.16 # hey, some problem with the registerform? can`t register because of the verification question. my answer is correct. 17.01.13 # you type "wombat" 17.01.33 # so if it says (what is 2+2), just type wombat 17.02.09 # lol..ok... :D 17.02.31 # to easy.. ;) 17.02.37 # (not for me) 17.02.48 # thanks! 17.03.47 # How do I add a DSP? What functions does it need? What functions can it access? 17.07.00 # Also, where/how does it access the PCM data? 17.08.58 Quit mortalis (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/) 17.10.58 # ah... I see 17.11.24 # You put "DSP_PROC_DB_ENTRY(..., DSPNAME_configure)" at the end 17.12.05 # and DSPNAME_configure takes a pointer to a dsp_proc_entry which has a function pointer to the function that does the actual processing 17.16.31 Quit krabador (Quit: Take the time.) 17.24.01 # TheSeven, I'm taking ploco's advice and am going to add a DSP that plays the audio through the piezo 17.29.22 # doesn't have to be at the end. it should be in the best place in the order. things are applied in the listed order 17.29.43 # oh hi jhMikeS 17.29.52 # (btw, you're marked as away) 17.30.00 # nvm 17.30.35 # my script does that if I haven't typed anything for (iirc) 30 min 17.30.41 # ah 17.31.03 # ok, for now, I just want a DSP that does nothing 17.31.43 # b 17.31.45 # sorry 17.31.48 # there's not much to that then 17.31.57 # but I want to be able to expand it later 17.32.07 # It's going to play PCM audio through the piezo :) 17.32.33 # several seconds ahead of the headphones :) 17.32.43 # what? 17.33.15 # actually, you won't get a good result because that empties and fills the buffer at codec speed, not PCM speed 17.34.01 # and it's buffered ahead of what you hear 17.34.13 # well, if I don't use headphones, it won't matter :) 17.34.33 # and anyway, I'm playing music through the piezo... so I'm not expecting a good result :) 17.35.03 # see my first remark, it'll speed up and slow down the audio if the buffer is draining then refilling, not just pegged 17.35.14 # ah 17.35.34 # then a DSP wouldn't be a good way to do it? 17.35.42 # no 17.35.52 # then what would? 17.36.09 # A tsr plugin that uses the mixer buffer capture might be better 17.36.29 # well, that's in my comfort zone :) 17.37.13 # and it would let the user start/stop easily 17.37.20 # oscilloscope uses the buffer capture callback (only one callback at a time per channel right now) 17.37.49 # so, if it's a plugin, they won't step on each other's toes 17.38.11 # clarify please? 17.39.06 # the playback audio channel can only have one callback to capture the audio at a time 17.39.51 # a tsr plugin would have to exit before loading a different one, which should clear the cb before exiting 17.40.24 # ah 17.40.29 # ok 17.42.10 # that's a good thing... not everyone wants music through the piezo, and that would be easier for me to add than, say, a settings option 17.42.26 # hmm... what should it be called? 17.44.52 # crystal meth? 17.46.06 Join krabador [0] (~krabador_@unaffiliated/krabador) 17.47.56 # rockcrystal 17.55.30 # rockrystal 17.55.44 # rocquartx 17.55.46 # rocquartz 18.01.29 # piezoplay? 18.06.37 # realsound? 18.12.05 # jackhammer 18.12.21 # cause it'll probably sound about as good :) 18.12.24 # haha 18.12.46 # gevaerts suggested that droppiing an iPod would make it sound better than the piezo would... 18.12.58 # "freefall"? 18.13.46 *** Saving seen data "./dancer.seen" 18.14.40 # will in it play in stereo if you use 3-D earmuffs? 18.16.15 #