| 00:56:14 | | Quit pixelma (Quit: .) |
| 00:56:15 | | Quit amiconn (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) |
| 00:57:14 | | Join amiconn [0] (jens@p4fd7f6f5.dip0.t-ipconnect.de) |
| 00:57:15 | | Join pixelma [0] (marianne@p4fd7f6f5.dip0.t-ipconnect.de) |
| 01:00 |
| 01:01:02 | | Quit pixelma (Client Quit) |
| 01:01:02 | | Quit amiconn (Client Quit) |
| 01:07:55 | | Join pixelma [0] (marianne@p200300ea87462b00305e95fffec66ff3.dip0.t-ipconnect.de) |
| 01:07:55 | | Join amiconn [0] (jens@p200300ea87462b00305e95fffec66ff3.dip0.t-ipconnect.de) |
| 01:30:23 | *** | Saving seen data "./dancer.seen" |
| 02:00 |
| 02:08:27 | edhelas_ | speachy From what I see, all the translations have a "(translated Track):" as a translation, so we might be able to simply remove the ":" and put it back directly in the code |
| 02:08:59 | edhelas_ | Same for the others, this would allow to translate a few more strings in the UI |
| 02:09:11 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 02:13:42 | | Quit dconrad (Ping timeout: 252 seconds) |
| 02:15:38 | edhelas_ | > if a string in tagnavi.config exists in the master language file, it will get used |
| 02:15:39 | edhelas_ | That's actually not the case looks like. I do have a French translation for "Artist" (LANG_ID3_ARTIST) which is used in the following line in tagnavi.config but not translated in the end |
| 02:15:39 | edhelas_ | "Artist" -> canonicalartist -> album -> title = "fmt_title" |
| 03:00 |
| 03:30:27 | *** | Saving seen data "./dancer.seen" |
| 03:50:14 | | Quit advcomp2019_ (Read error: Connection reset by peer) |
| 04:00 |
| 04:44:13 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 04:48:59 | | Quit dconrad (Ping timeout: 272 seconds) |
| 05:00 |
| 05:03:59 | | Quit jacobk (Ping timeout: 260 seconds) |
| 05:04:45 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 05:30:30 | *** | Saving seen data "./dancer.seen" |
| 07:00 |
| 07:18:32 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 07:22:57 | | Quit dconrad (Ping timeout: 246 seconds) |
| 07:30:31 | *** | Saving seen data "./dancer.seen" |
| 07:42:06 | | Quit decky_e (Read error: Connection reset by peer) |
| 08:00 |
| 08:26:12 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 08:49:37 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 09:00 |
| 09:30:35 | *** | Saving seen data "./dancer.seen" |
| 10:00 |
| 10:02:39 | speachy | dconrad: IIRC we currently have the erosq (and erosq_h2), agptekrocker, xduoox3ii, and xduoox20 as our hibyos version.txt targets. |
| 10:03:43 | speachy | edhelas_: the ':' is intentionally there; those strings are used as list items so the code that uses them would need to be suitably corrected. |
| 10:04:05 | speachy | (basically that's why I didn't just change that stuff a while back..) |
| 10:17:10 | | Join macaronus [0] (~macaronus@user/MaCaRoNus) |
| 10:18:26 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 10:28:19 | edhelas_ | I see |
| 10:33:41 | speachy | In other contexts we've been getting rid of punctuation in translated strings |
| 10:34:18 | edhelas_ | That's a good thing ! |
| 10:35:24 | | Quit othello7 (Ping timeout: 245 seconds) |
| 10:35:48 | speachy | it's a process. :) |
| 10:36:24 | speachy | patches welcome, as always |
| 10:36:53 | edhelas_ | By the way, do you have specific tasks/todo list regarding Rockbox, always curious to knows "whats planned next" |
| 10:38:13 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 10:38:49 | speachy | There's not any sort of public/shared roadmap, if that's what you mean. Everyone [semi-]active has their own personal priorities. |
| 10:39:54 | speachy | I've been trying to consistently put "concrete" problems into the issue tracker if it's not something I can knock out quickly. |
| 10:42:35 | | Quit dconrad (Ping timeout: 255 seconds) |
| 11:00 |
| 11:11:13 | | Join advcomp2019 [0] (~advcomp20@user/advcomp2019) |
| 11:19:00 | user890104 | edhelas_: i'm trying to get ramdisk working, but hit a dead end in the usb stack. i need it for an early port to iPod nano 3g and 4g |
| 11:26:40 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 11:27:08 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 11:28:41 | OlsroFR | About me personnally, I am now starting using more my iPods and am very satisfied by how Rockbox performs on my pods. I just feel like it is laggy on my iPod Video, like sometimes even more laggy compared to how it runs on my iPod Mini or on my iPod 4th gen mono. It's strange; probably related by the size of the 320x240 color screen that is more |
| 11:28:41 | OlsroFR | difficult to handle by Rockbox ? |
| 11:29:21 | | Quit paulk (Ping timeout: 246 seconds) |
| 11:29:45 | | Join paulk [0] (~paulk@about/aquilenet/user/paulk) |
| 11:29:58 | OlsroFR | It's not unusable or very laggy, but it is a bit sluggish. |
| 11:30:20 | OlsroFR | Storage is also a real compact flash so no difference with my Mini |
| 11:30:23 | speachy | same processor/busses several times the amount of data to sling around for each display update |
| 11:30:36 | *** | Saving seen data "./dancer.seen" |
| 11:31:06 | OlsroFR | even more strangely, my ipod mini is very very slow at building the database with the same amount of songs. Almost 1 hour. But my ipod classics (the 4th gen and the video) can build it much much faster, something like 2x to 3x faster |
| 11:31:30 | | Quit dconrad (Ping timeout: 276 seconds) |
| 11:32:35 | OlsroFR | the mini should have the same power as a full sized ipod so it is strange, but maybe it has a slower bus/disk accesses ? |
| 11:32:42 | speachy | identical storage? |
| 11:33:19 | OlsroFR | The mini has a compact flash from amazon, a 128GB one. This one https://www.amazon.fr/gp/product/B00NUB2TWI |
| 11:33:48 | OlsroFR | the 2 other pods have both a Kingspec 256GB from Aliexpress. That famous cheap CF card... |
| 11:33:52 | speachy | the 4g is equivalent to the mini2g, |
| 11:34:04 | speachy | the min1g has a pp5020 vs the 5022 of the later ones |
| 11:35:07 | OlsroFR | it's strange to get a db that builds so slow especially on a CF card that was so expensive. For the price of that 128GB Sandisk CF card, it is possible to get 2x256GB ones from Aliexpress. The only difference is that delivery from Aliexpress takes 3 weeks... |
| 11:35:09 | speachy | offhand I couldn't tell you what the differences are. |
| 11:35:52 | speachy | you could run a disk benchmark on each to see what the relative differences are. |
| 11:36:00 | OlsroFR | Good idea. I will do it now |
| 11:37:28 | OlsroFR | How can I do this ? |
| 11:38:22 | speachy | you'll need to do an advanced build with the Test plugins, IIRC |
| 11:38:39 | OlsroFR | Ah :( |
| 11:40:02 | speachy | that'll generate a decent pile of test plugins, notably including the codec and disk benchmarks. |
| 11:40:06 | OlsroFR | anyway, even if on the mini the db takes forever to build, after it's done the experience is very fast and smooth. Smoother than the one on my ipod video |
| 11:41:14 | OlsroFR | when I click play, my ipod video freezes during around 3 seconds to load the song and its cover art. It seems like showing the cover art has a really huge impact on performance on my ipod video |
| 11:41:29 | OlsroFR | if I disable cover arts completely, it's much faster |
| 11:41:44 | OlsroFR | all my cover arts are external files .jpg at 200x200 so it's not huge |
| 11:42:19 | OlsroFR | I am wondering if converting to bmp will lead to improvement regarding this issue because bmp is uncompressed |
| 11:42:20 | speachy | jpeg decoding (and scaling) is expensive. |
| 11:43:59 | OlsroFR | But for some reason on my ipod mini/4th gen I see this issue much less. Probably because they have a tiny black&white screen ? I don't know. But they have a much less overhead when trying to show my jpegs |
| 11:44:35 | OlsroFR | edhelas_ right now I do not have any new patch in development |
| 12:00 |
| 12:12:03 | | Quit thanosengine (Ping timeout: 246 seconds) |
| 12:14:17 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 12:16:45 | edhelas_ | speachy I saw some assembly code regarding JPEG in the code, its that expensive ? |
| 12:17:54 | | Quit zem (Quit: and it's always been the same / it's just a complicated game) |
| 12:18:19 | | Quit thanosengine (Read error: Connection reset by peer) |
| 12:18:25 | | Join hactar|ant [0] (~zem@97-115-91-140.ptld.qwest.net) |
| 12:18:58 | hactar|ant | olsrofr how many songs do you have on the ipods? |
| 12:19:29 | OlsroFR | hactarland Around 30000 songs on the classics, and around 27000 on the mini |
| 12:19:52 | OlsroFR | On the mini, all songs are converted with AAC144kbps to fit better in the reduced storage amount. On the big ipods, it's musepack q7 |
| 12:19:58 | hactar|ant | oh wow ok |
| 12:20:31 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 12:22:07 | hactar|ant | i was going to say, i have rockbox on ipod 4g, 5.5g, and mini 2g, and the 5.5 is fastest to build the database, but there isn't a noticeable difference between the 4g and m2g |
| 12:22:14 | hactar|ant | but my libraries are much smaller |
| 12:22:49 | hactar|ant | the 5.5g has roughly 12000 tracks, 4g and m2g about 8000 |
| 12:23:34 | hactar|ant | i definitely don't notice a giant difference in ui responsiveness between any of them, unless i'm scrolling a list of hundreds of songs |
| 12:27:16 | | Join othello7 [0] (~Thunderbi@pool-100-36-176-164.washdc.fios.verizon.net) |
| 12:34:54 | | Quit jacobk (Ping timeout: 260 seconds) |
| 12:40:50 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 12:47:31 | speachy | edhelas_: the short answer is yes. at least relative to the raw capabilities of the ipods. |
| 12:50:50 | | Join Zeke [0] (~Zeke@226.219.9.51.dyn.plus.net) |
| 12:51:34 | | Quit Zeke (Client Quit) |
| 12:54:50 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 13:00 |
| 13:10:55 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 13:27:39 | | Quit macaronus (Ping timeout: 246 seconds) |
| 13:27:59 | | Join lebellium [0] (~lebellium@2a01cb0405d07f0035d7b2fde936a0d1.ipv6.abo.wanadoo.fr) |
| 13:28:01 | | Join IPG [0] (~InvoxiPla@94.4.117.11) |
| 13:30:40 | *** | Saving seen data "./dancer.seen" |
| 13:43:15 | | Join jacobk [0] (~quassel@utdpat242097.utdallas.edu) |
| 13:44:18 | OlsroFR | I tried to convert all my jpeg cover art to bmp. It seems faster/smoother to load overall |
| 13:48:27 | edhelas_ | OlsroFR the python script you have is actually converting progressive jpeg to non-progressive ? |
| 13:48:48 | OlsroFR | I am not using it anymore but yeah, it was doing real jpeg. |
| 13:48:56 | OlsroFR | I am now using the tool "mp3tag" |
| 13:49:12 | OlsroFR | and now I use also imagemagick in cli to convert .jpg to bmp |
| 13:49:24 | OlsroFR | I use 2 clis to do it recursively : |
| 13:49:26 | OlsroFR | find . -name '*.jpg' -type f -exec mogrify -format bmp {} + |
| 13:49:29 | OlsroFR | find . -name '*.jpg' -type f -delete |
| 13:49:34 | | Quit jacobk (Ping timeout: 260 seconds) |
| 14:00 |
| 14:16:24 | | Quit OlsroFR (Quit: Client closed) |
| 14:22:33 | | Join davisr [0] (~davisr@fsf/emeritus/davisr) |
| 14:46:09 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 14:50:30 | | Quit dconrad (Ping timeout: 252 seconds) |
| 14:57:26 | | Join jacobk [0] (~quassel@utdpat242064.utdallas.edu) |
| 15:00 |
| 15:05:03 | | Join thanosengine [0] (~thanos@user/thanosengine) |
| 15:10:39 | | Quit jacobk (Ping timeout: 260 seconds) |
| 15:30:41 | *** | Saving seen data "./dancer.seen" |
| 15:35:41 | | Quit Guest2797 (Changing host) |
| 15:35:41 | | Join Guest2797 [0] (sid23779@fsf/member/yang) |
| 15:36:42 | | Nick Guest2797 is now known as yang2 (sid23779@fsf/member/yang) |
| 15:46:40 | | Quit thanosengine (Quit: WeeChat 4.3.1) |
| 16:00 |
| 16:12:30 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 16:12:35 | OlsroFR | that new version of Gerrit with dark mode is so cool |
| 16:15:08 | speachy | it had a dark mode before |
| 16:15:24 | OlsroFR | it was not enabled on me. Now it is |
| 16:15:41 | OlsroFR | Probably because it can now detect the dark mode parameter system wide. On my mac, I enabled dark mode everywhere |
| 16:15:54 | OlsroFR | and some sites are clever enough to automatically adapt to the system settings ;) |
| 16:16:21 | speachy | yeah, it had to be done manually bedore |
| 16:22:50 | dook | OlsroFR re: sluggish performance. If you're using a theme with CJK support, it could be a cause of lag. I've gotten more than a few reports about it. |
| 16:23:37 | OlsroFR | dook I noticed that playing japanese songs seemed to add lags yeah sometimes. It's like rendering symbols can be difficult sometimes |
| 16:25:03 | OlsroFR | Sazanami mincho font that I use with my ipod mini/4th gen mono does not exactly have perfect support for all symbols, so it's maybe more lite than a full featured font |
| 16:25:27 | OlsroFR | on my ipod video, I use my fixed version of the theme ipod3blk which use Cantarell-18 (the font that you also use for adwaitapod) |
| 16:27:47 | dook | Ah yeah there you go, I use noto sans like freshos which is 40,000 characters or something crazy |
| 16:28:39 | dook | I want to make an auto subsetting tool like those for websites at some point |
| 16:30:53 | OlsroFR | This would be great to maybe cache the fonts in RAM fully to avoid those ugly lags in rendering ? |
| 16:31:09 | OlsroFR | I don't know how this really works, but maybe the Rockbox OS read the whole font files each time it need to render characters |
| 16:31:55 | OlsroFR | i would gladly sacrifice some very little RAM from my 64MB iPod Video to cut lags |
| 16:34:15 | dook | I think Rockbox caches fonts? Can't remember though |
| 16:34:19 | OlsroFR | i tried to switch back to the Cabbie theme and it's not so much faster overall |
| 16:35:40 | dook | Actually yeah, it must. I've ran into some strange behaviour with fonts that are easiest explained by them being cached |
| 16:55:39 | speachy | rb caches font glyphs, yes. |
| 16:56:45 | OlsroFR | +speachy Does it cache font glyphs instantly or with some lazy caching ? |
| 16:57:11 | OlsroFR | (what I mean is that it will cache glyphs one by one when those are seen for the first time) |
| 17:00 |
| 17:04:46 | | Quit kugel_ (Quit: Lost terminal) |
| 17:06:25 | speachy | this is just speculation, but I would expect it will only cache things as-requested, and keep them around in an LRU fashion. |
| 17:09:39 | | Join kugel [0] (~kugel@ip4d146a3a.dynamic.kabel-deutschland.de) |
| 17:18:06 | | Quit OlsroFR (Quit: Client closed) |
| 17:22:00 | gevaerts | I *think* the glyph cache is an LRU cache, yes |
| 17:23:25 | speachy | it's not really feasible to cache entire fonts. Each display element can have its own font, and each of those could theoretically contain a full set of codepoints. |
| 17:24:49 | edhelas_ | So it seems that the skinned_menu engine is actually used by some themes :) ! https://github.com/D0-0K/adwaitapod/issues/23#issuecomment-2344568957 |
| 17:25:41 | gevaerts | Fonts can be *massive* too |
| 17:25:49 | gevaerts | As in, several times the RAM of some players |
| 17:27:13 | speachy | IIRC unicode font makes up something like half the flash image on the hibyos-based players. |
| 17:27:24 | speachy | (ie 20+MB, and that's just one font) |
| 17:30:22 | gevaerts | The unifont as packaged in the rockbox font pack seems to be 2MB |
| 17:30:44 | *** | Saving seen data "./dancer.seen" |
| 17:31:17 | speachy | yeah, was just checking on that. >2MB for the 16px unifont. |
| 17:33:57 | | Quit lebellium (Quit: Leaving) |
| 17:36:36 | gevaerts | Of course if the cache is too small things are not going to work very fast... |
| 17:36:57 | speachy | depends on how many unique glpyhs you have on-screen at once |
| 17:37:50 | speachy | isn't usually a problem unless you _also_ have multiple fonts in use at once. |
| 18:00 |
| 18:05:58 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 18:10:23 | | Quit dconrad (Ping timeout: 255 seconds) |
| 18:20:20 | | Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net) |
| 18:57:47 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 19:00 |
| 19:00:12 | OlsroFR | Well, lags are not terrible, really. I enjoyed using my ipod video several hours today. It was good enough to be a good experience. I wished it could be slightly faster sometimes but it's not so bad. Anyway I feel like fonts is not the main issue of why the ipod video is noticeably slower than my grayscale ipods. It's probably just by the fact that |
| 19:00:12 | OlsroFR | it has the same hardware of older ipods but with a bigger screen |
| 19:01:05 | | Quit OlsroFR (Client Quit) |
| 19:03:16 | speachy | it's ~16x the amount of data being shoveled around for the display (~4x the number of pixels and 4x the per-pixel data) |
| 19:05:24 | speachy | IIRC the display framebuffer is also in IRAM, which leaves less room for other stuff (eg code) |
| 19:19:48 | | Quit kugel (Ping timeout: 244 seconds) |
| 19:30:46 | *** | Saving seen data "./dancer.seen" |
| 19:34:37 | _bilgus | We have a setting 'Glyphs To Cache' Settings>General Settings>System>Limits>Glyphs To Cache does raising or lowering make the problem better/worse? |
| 19:41:35 | | Join kugel_ [0] (~kugel@ip4d146a3a.dynamic.kabel-deutschland.de) |
| 19:42:43 | | Join massiveH [0] (~massiveH@2600:4040:a982:dc00:8c6c:9ea0:6838:b492) |
| 19:54:09 | | Join jacobk [0] (~quassel@2603:8080:b200:7b02:77cb:6304:f9db:dda1) |
| 20:00 |
| 20:00:35 | | Quit davisr (Quit: yeehaw) |
| 20:01:06 | | Quit jacobk (Ping timeout: 276 seconds) |
| 20:05:27 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 20:09:55 | | Quit dconrad (Ping timeout: 244 seconds) |
| 20:10:35 | | Quit Galois (Remote host closed the connection) |
| 20:31:51 | | Join Galois [0] (djao@efnet.math.uwaterloo.ca) |
| 21:00 |
| 21:18:56 | | Join jacobk [0] (~quassel@47-186-105-237.dlls.tx.frontiernet.net) |
| 21:30:49 | *** | Saving seen data "./dancer.seen" |
| 22:00 |
| 22:20:42 | | Quit Moriar (Ping timeout: 246 seconds) |
| 22:42:50 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 22:47:10 | | Quit dconrad (Ping timeout: 252 seconds) |
| 22:47:43 | | Quit massiveH (Quit: Leaving) |
| 22:57:44 | | Join dconrad [0] (~dconrad@152.117.104.217) |
| 23:00 |
| 23:18:41 | _bilgus | anyone care to look at g#5926 and see if there is anything else I might have missed? |
| 23:18:45 | rb-bluebot | Gerrit review #5926 at https://gerrit.rockbox.org/r/c/rockbox/+/5926 : RFC skin_tokens remove a static buffer by William Wilgus |
| 23:30:52 | *** | Saving seen data "./dancer.seen" |
| 23:40:13 | | Quit othello7 (Remote host closed the connection) |
icon identifies that the person is a core developer (has commit access).