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 2025-02-26

00:03:59 Join chillwat [0] (~chillwat@c-174-53-199-178.hsd1.mn.comcast.net)
00:35:10***Saving seen data "./dancer.seen"
02:00
02:08:15 Quit chillwat (Read error: Connection reset by peer)
02:35:11***Saving seen data "./dancer.seen"
02:52:57 Quit thanosengine (Quit: WeeChat 4.5.1)
02:55:02 Join thanosengine [0] (~thanos@user/thanosengine)
02:55:32 Quit COMPL_EXE (Read error: Connection reset by peer)
02:55:50 Join COMPL_EXE [0] (~compl.exe@aosc/dev/origincode)
03:00
03:07:25 Join advcomp2019_ [0] (~advcomp20@user/advcomp2019)
03:10:54 Quit advcomp2019 (Ping timeout: 268 seconds)
03:22:15 Quit b0 (Quit: Ping timeout (120 seconds))
03:22:28 Join b0 [0] (~b0@user/b0)
03:55:28 Quit othello7 (Ping timeout: 252 seconds)
04:00
04:04:18 Quit jacobk (Ping timeout: 248 seconds)
04:04:39 Join jacobk [0] (~quassel@47-186-65-73.dlls.tx.frontiernet.net)
04:35:14***Saving seen data "./dancer.seen"
04:54:03 Quit Malinux (Ping timeout: 252 seconds)
04:58:06 Quit jn ()
04:58:23 Join jn [0] (~quassel@user/jn/x-3390946)
05:00
05:04:00 Quit acidsys (Ping timeout: 244 seconds)
05:20:31 Join acidsys [0] (~crameleon@openSUSE/member/crameleon)
05:56:57 Quit COMPL_EXE (Ping timeout: 265 seconds)
05:56:59 Quit PheralSparky (Read error: Connection reset by peer)
05:57:22 Join COMPL_EXE [0] (~compl.exe@aosc/dev/origincode)
05:59:46 Quit emash (Ping timeout: 252 seconds)
06:00
06:01:44 Join emash [0] (~emash@user/emash)
06:03:28 Join Everything [0] (~Everythin@46.211.65.235)
06:31:01 Quit Everything (Ping timeout: 248 seconds)
06:31:50 Quit jn (Ping timeout: 244 seconds)
06:32:05 Join jn [0] (~quassel@2a0a-a549-e589-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de)
06:32:06 Quit jn (Changing host)
06:32:06 Join jn [0] (~quassel@user/jn/x-3390946)
06:35:18***Saving seen data "./dancer.seen"
07:00
07:47:12_bilgus_well at default the HM800 user says its stable, thats good to hear, maybe we have some wonky settings happening..
08:00
08:28:57speachybleb: "Easily replaceable battery" is the tough one.
08:29:39speachythe modern erosq family is relatively easy to take apart and do a battery swap (screws but they're under glued-on covers)
08:30:36speachybut the only thing I"m aware of with an easily replaceable battery are the iBasso DX50/DX90 models. they have an explicitly removeable battery, IIRC one that samsung used to use for their phones a long time ago
08:31:09speachy_bilgus_: ...I get the impression there is a lot we're not being told.
08:35:19***No seen item changed, no save performed.
08:36:25_bilgus_language barrier I think is most of the issue, hopefully we get some bugfixes out of it
08:38:03_bilgus_If I'm not mistaken I think we had one device that used Alkaline AAs maybe hwcodec?
08:40:26speachythe OG archos devices uses 4x AA batteries, came with NiMH cells but you could swap in whatever you wanted
08:40:34speachywe put in discharge curves for alkalines too
08:43:43speachythere were some other players in the early era that had AAs (Creative had one that looked like a USB storage stick but plugged into a battery caddy that contained a single AA
08:45:36speachyI think that line topped out at 1GB.
08:49:32speachywrt the HM801, the device is on the underpowered side (or at least the pcm code could be better optimized)
08:49:52speachywas pretty easy to get the output to crackle a bit from buffer underruns
08:50:49_bilgus_so adding any extra processing on the audio will probably bring up so,e stuff
08:50:58_bilgus_some*
08:51:56_bilgus_yeah the archos I think it was the first device I saw RB on
08:52:47speachyI was able to make it crackle playing back 2ch 44khz FLAC, mucking around in the debug menu
08:53:42speachyplenty of spare CPU cycles (I think it was effectively running at 1/4 max CPU)
08:54:08speachyMaybe it's a matter of making the PCM DMA buffer slightly larger
08:55:02speachya few months ago I found the documentation and software CD for my long-deceased Archos Studio
08:55:16_bilgus_yeah buy you a bit more time< usually though that just moves the issue a bit further down the line but maybe its that close?
08:56:39speachydepends on the root of the problem −− eg the jz47xx has plenty of oomph but the ISR overhead is huge and the dma engine doesn't let you do double-buffering
08:57:13speachyso we get underruns at very high rates solely because we can't service the ISR fast enough.
08:58:13speachy(the context switch is _expensive_)
08:58:27_bilgus_ah so extend taht out and get more favorable timings huh
08:59:06speachybut on the rk27xx it might just be a matter of the buffers not being deep enough to refresh in time.
08:59:21_bilgus_and minimal testing doesn't help either
08:59:43speachyyeah.
09:00
09:02:01speachy(on the jz47xx we're saved by a relatively deep i2s FIFO. the x1000 series added end-of-descriptor IRQs instead of just end-of-transfer, so we can double-buffer. plus another 400-500MHz of oomph..)
09:02:54_bilgus_We have cleaned up a lot of this common code to the point where our major issues should be relegated to single devices and dark corners but those are also usually the hardest to track down too
09:03:12_bilgus_esp without a device to test..
09:04:00speachyI expect the HM601 is effectively equivalent.
09:04:16speachybut dma underruns wouldn't cause panics
09:05:13_bilgus_overclock mods wouldthough..
09:05:55CtcpIgnored 4 channel CTCP requests in 12 days and 22 hours at the last flood
09:05:55*speachy spreads his hands.
09:07:13_bilgus_I asked them again what mods
09:08:47speachyI still need to finish generating a flashable bootloader image on the DX50. So far it won't boot.
09:09:10_bilgus_fallout from the compiler jump?
09:09:23speachyoh no, this is just reliant on dodgy tools
09:09:52_bilgus_heh
09:10:06speachythe open source tools don't work properly for this SoC and I wasn't motivated to hunt down the OG proprietary vendor tooling
09:10:59speachy(basically need to generate a flashable/bootable image after patching the hosted filesystem with our bootloader)
09:11:55speachy (it's all hosted and compiled with stock android toolchains so I don't expect any meaningful issues in that respect)
09:14:02speachystill need to regenerate a new set of bootloaders for the ipods and everything that takes sd cards.
09:14:20speachy(ie to pull in the disk/partitioning improvements)
09:16:20_bilgus_yeah I haven't tested any of the new bootloaders except the clip zip I think the work amachronic did to make stuff fit last time (for clip+ and...?) left some room on the table for you so hopefully it fits
09:16:42speachywe build all bootloaders after every commit
09:16:58_bilgus_oh so they fit then they just need tested?
09:17:03speachyyeah
09:17:08_bilgus_cool
09:17:51_bilgus_that is a much better state then, now do you mean they are fed through the makegen too?
09:18:16speachymakegen?
09:18:18_bilgus_or just the first stage bc those don't check if it fits in the OF
09:18:45_bilgus_like for the ams devices you use makeamsboot to merge the OF and our bootloader
09:18:55speachyno, we dont' do that
09:19:04_bilgus_then we can't say it fits
09:19:26speachybut if there's a size limit for the bootloader it should be referenced in the boot.lds for that target
09:19:35_bilgus_I doubt it
09:19:42_bilgus_hope so?
09:19:49speachy"should" :D
09:20:10speachyI know that in the past we've had buildfarm failures along those lines, so at least _some_ targets do the right hting
09:20:30speachywhat's one of the problematic ones?
09:21:08_bilgus_for sure the clip+ has been an ongoing pain
09:21:46_bilgus_there are others that were close but off the top of my head clip+
09:22:16speachyok, that just uses DRAM : ORIGIN = DRAM_ORIG + LOAD_SIZE, LENGTH = DRAM_SIZE - LOAD_SIZE - TTB_SIZ
09:22:20_bilgus_last time around I think amachronic managed something like 1k free
09:22:40_bilgus_give me a second and i'll spin up a makeamsboot for it
09:23:32speachyif we overflow I'd like to start tweaking the linker scripts to reflect the actual limit
09:29:08_bilgus_fits by 3k I think you'll be fine
09:29:49_bilgus_well 2,4k
09:30:45_bilgus_[INFO] Original firmware size: 134008 bytes, [INFO] Total size of new image: 131564 bytes
09:31:13speachyheh but does it _work_ ? :D
09:31:20_bilgus_I don't know that you could, I guess since we have no ned OF to contend with we know what all the possible sizes are
09:31:31_bilgus_no new*
09:31:48speachyI figure we only care about the final firmware release
09:32:12_bilgus_and or whatever is the LARGEST
09:48:59_bilgus_The Sansa Clip+ Does Boot
09:57:55_bilgus_Hm800 sounds like the mods are in the audio path so power dropouts maybe but doubtful to cause much issue otherwise I'd think
10:00
10:02:02_bilgus_I tell you I love the Clip+
10:02:49_bilgus_<3 <3 <3
10:02:56speachyyeah, it's an amazing little device. my only complaint is that the screen is too teeeeny for my getting older eyes now.
10:03:07speachy(well, that and the lack of a line out, but that's understandable)
10:03:19_bilgus_I just don't care TBH i'll use it with voice damnit
10:04:34_bilgus_but yeah my eyes aren't getting beter, I'm glad chris_s added zoom option on te sim but it just makes me feel old now
10:14:38speachyvoice is what makes the x3 usable too. plus the big chonky buttons that are easily distinguishable by touch
10:21:59 Join ac_laptop [0] (~ac_laptop@2a0c:e303:0:7300:e29d:31ff:fe2d:a258)
10:24:46Xogiumno shame in having poor eyesight, honest folks
10:25:17speachythe main benefit for the new bootloader is GPT support for folks that reformat but not repartition their SD cards.
10:25:35speachy(for SD card devices)
10:26:19speachyand yeah, the voiced UI remains rockbox's killer feature
10:26:26Xogiumaye, that it does
10:28:02XogiumI wrote a guide on how to use the zoom f3 field recorder device while blind, and was pleasantly surprised to read that someone posted a comment that said, thanks! This will be bloody useful to me when I want to perform some task on the recorder without glancing at the display
10:28:39Xogiumsometimes you just don't realize how good something you've just decided to make for your own use can be to others until they point it out and you're like, huh wow that's nice
10:35:20***No seen item changed, no save performed.
10:40:31speachyit's why I've put so much effort into making it (and the tooling) JustWork(tm).
10:41:37 Quit _bilgus_ (Ping timeout: 248 seconds)
10:43:16Xogium:)
10:43:33Xogiumdamn fine job you did for sure
10:47:42speachy"if I have seen further it is because I am standing on the shoulders of giants"
10:50:16 Join _bilgus [0] (~bilgus@syn-162-154-213-134.res.spectrum.com)
11:00
11:08:48 Quit jjs0077018310196 (Ping timeout: 268 seconds)
11:18:42 Join jjs0077018310196 [0] (~jjs007@host86-191-58-228.range86-191.btcentralplus.com)
11:21:01speachy_bilgus: https://www.ebay.com/itm/275810517140
11:23:32 Quit jjs0077018310196 (Client Quit)
11:24:08_bilgushmm
11:24:49_bilgusI could justify 300 1200 is a bit much for something redundant in my life lol
11:26:09 Join jjs0077018310196 [0] (~jjs007@host86-191-58-228.range86-191.btcentralplus.com)
11:38:23 Join JanC_ [0] (~janc@user/janc)
11:38:23 Nick JanC is now known as Guest5429 (~janc@user/janc)
11:38:23 Quit Guest5429 (Killed (silver.libera.chat (Nickname regained by services)))
11:38:23 Nick JanC_ is now known as JanC (~janc@user/janc)
11:42:58 Join JanC_ [0] (~janc@user/janc)
11:42:58 Nick JanC is now known as Guest4363 (~janc@user/janc)
11:42:58 Quit Guest4363 (Killed (calcium.libera.chat (Nickname regained by services)))
11:42:58 Nick JanC_ is now known as JanC (~janc@user/janc)
11:47:51 Join bahusoid [0] (~bahusoid@31.43.34.240)
11:52:35bahusoidI think e200V2 bootloader is the most size-restricted. I worked on fitting it and if I remember correctly I needed to clean up 12Kb or something.
11:52:46bahusoidhttps://forums.rockbox.org/index.php/topic,54768.0.html
11:55:01bahusoidAlso, I don't believe that the bootloader build checks if it fits in firmware as it requires original firmware. Otherwise why e200v2 bootloader was broken for many years
11:56:48bahusoidAnd I verified that GPT partition works for e200v2 bootloader
11:59:07 Quit bahusoid (Quit: Client closed)
12:00
12:11:48speachythose limitations should be encoded into the linker scripts
12:12:15speachy(or at minimum, in the configure script)
12:18:55_bilgusI think at the very least you'd need o run it through the compressor first UCLpack
12:19:16_bilgusas we can't say what the compressed size will be ahead of time more than a guess
12:22:09 Quit jacobk (Ping timeout: 260 seconds)
12:23:55blebspeachy: thanks for the run down
12:24:11blebthe archos recorders look interesting
12:24:31bleb<speachy> there were some other players in the early era that had AAs (Creative had one that looked like a USB storage stick but plugged into a battery caddy that contained a single AA
12:24:48blebI've spent a while researching this type; I think the creative one you had in mind used AAA
12:25:31_bilgusclip plus gets about 40% smaller
12:25:43blebAA ones are less common, the one's I've found are cowon g3, samsung yepp mt6, and a bunch of irivers
12:26:13blebafaik none of them will charge the battery though so you have to constantly swap it out into a recharger
12:26:21blebnot too bad since it lasts a while
12:26:48_bilgusthere was one I reember maybe creative that was a small circle with a button and had a screen off the side O== taht took two AAA
12:27:26blebthe archos players and also creative jukeboxes that took 4 AAs came with NiMH and had circuitry to charge them
12:27:32_bilgusit was actually a ring of buttons
12:29:46 Join GreenYoshi [0] (~GreenYosh@user/GreenYoshi)
12:29:52blebI would like to find that
12:30:26blebone AAA seems a little weak but two AAA could be alright
12:31:44_bilgushttps://www.newegg.com/sandisk-sansa-m240-mp3-player/p/N82E16855125117
12:32:34_bilguswas one AAA
12:32:43blebyup
12:32:44_bilgusapparently
12:34:22 Quit GreenYoshi (Client Quit)
12:34:26blebthose do look really nice though
12:35:22***Saving seen data "./dancer.seen"
12:35:38_bilgusI think they expected you to swing it in your hand while running or something
12:36:44 Join JanC_ [0] (~janc@user/janc)
12:36:44 Quit JanC (Killed (lithium.libera.chat (Nickname regained by services)))
12:36:44 Nick JanC_ is now known as JanC (~janc@user/janc)
12:42:28blebsansa e200 seems to have everything but the port is proprietary
12:42:38bleband replacement batteries seem a bit scarce
12:43:01blebI guess the real answer is tangara
12:44:57 Join GreenYoshi [0] (~GreenYosh@user/GreenYoshi)
12:45:33 Quit GreenYoshi (Remote host closed the connection)
12:45:58 Join GreenYoshi [0] (~GreenYosh@user/GreenYoshi)
12:47:47 Quit GreenYoshi (Remote host closed the connection)
12:48:08 Join GreenYoshi [0] (~GreenYosh@user/GreenYoshi)
12:50:49 Join chillwat [0] (~chillwat@c-174-53-199-178.hsd1.mn.comcast.net)
12:51:36 Quit thanosengine (Quit: WeeChat 4.5.1)
12:53:32 Join lebellium [0] (~lebellium@2a01cb0405d07f0048de7585089e91dd.ipv6.abo.wanadoo.fr)
12:54:33 Quit GreenYoshi (Quit: Leaving)
12:55:49_bilgusyeah almost exactly the opposite of proprietary
12:56:22speachy...except for a large chunk of the software running on it
12:56:37blebbluetooth/wifi firmware?
12:56:55bleboh wow I actually have an archos recorder 20 on the way
12:57:02blebebay just didn't tell me my offer was accepted
12:57:29speachycommunication firmware and stack at minimum, but likely numerous otehr ESP32 blobs
12:58:04speachy(ESP32 provides a _lot_ of libraries for use with their hardare)
13:00
13:04:34speachy(there's also a bunch of stuff that lives in ROM −− IIRC that includes the USB stack)
13:09:57_bilgusspeachy https://www.ebay.com/itm/123773223461
13:10:27speachywow, an order of magnitude more expensive.
13:12:22 Join Everything [0] (~Everythin@46-133-48-141.mobile.vf-ua.net)
13:13:36_bilgusWonder how much they were new, damn I should have inversted in them I guess
13:14:02speachythe 801 is better than the 601 insofar as it has more buttons
13:14:16speachybut the layout is a little head-scratchy
13:14:37speachyhuge battery too but I imagine some of those analog boards are quite thirsty
13:15:02 Join Malinux [0] (~malin@2a01:799:f22:bb00::12c:c4a7)
13:19:53 Join PheralSparky [0] (~S|h|a|w|n@user/shawn/x-4432647)
13:27:51speachyI'm grateful we still have the likes of the erosq family; bucking the trend by providing an option between shovelware and "audiophile" devices.
13:30:03speachythen there's whatever TF this is.. https://www.ebay.com/itm/356569454513
13:35:30 Quit Everything (Quit: leaving)
13:37:28 Join Everything [0] (~Everythin@46-133-48-141.mobile.vf-ua.net)
13:37:51EverythingHi all. What are some modern players on market compatible with rockbox?
13:54:05gnafuEverything: I just picked up a Surfans F20 (Eros Q variant), and I'm very happy with it so far.
13:57:51Everythinggnafu: tnx
14:00
14:05:36 Quit LjL (Read error: Connection reset by peer)
14:06:26 Join LjL [0] (~ljl@user/ljl)
14:17:14 Quit jjs0077018310196 (Quit: The Lounge - https://thelounge.chat)
14:31:06 Join jjs0077018310196 [0] (~jjs007@host86-191-58-228.range86-191.btcentralplus.com)
14:35:02 Join jacobk [0] (~quassel@utdpat241106.utdallas.edu)
14:35:26***Saving seen data "./dancer.seen"
14:40:32 Join paulk-bis [0] (~paulk@vpn-0-22.aquilenet.fr)
14:41:23 Quit paulk (Read error: Connection reset by peer)
14:58:44 Join othello7 [0] (~Thunderbi@pool-100-36-176-164.washdc.fios.verizon.net)
15:00
15:03:35 Join Moriar [0] (~moriar@107-200-193-159.lightspeed.stlsmo.sbcglobal.net)
15:28:50 Quit jjs0077018310196 (Quit: The Lounge - https://thelounge.chat)
15:57:10Everything.quit
15:57:13Everythingsorry
15:57:15 Quit Everything (Quit: leaving)
15:58:26 Quit jacobk (Ping timeout: 248 seconds)
16:00
16:17:10 Join jjs0077018310196 [0] (~jjs007@host86-191-58-228.range86-191.btcentralplus.com)
16:32:04 Quit jjs0077018310196 (Quit: The Lounge - https://thelounge.chat)
16:35:30***Saving seen data "./dancer.seen"
16:41:40 Join jjs0077018310196 [0] (~jjs007@host86-191-58-228.range86-191.btcentralplus.com)
17:00
17:12:40 Quit lebellium (Quit: Leaving)
17:33:40blebanyone know where I might find specs for the archos recorder 20 ac adapter?
17:34:38blebaccording to this the PJ-007 jack can be used as a replacement for the power jack: http://www.woudy.org/archos_apart/
17:35:18blebwhich is 1.3 x 4.2 mm, 2.0 A:
17:35:21blebhttps://www.sameskydevices.com/product/interconnect/connectors/dc-power-connectors/jacks/pj-007
17:36:21blebit's "rated" for 16V but idk what voltage the device is designed to accept
17:37:38speachypretty sure it's 5V
17:40:48speachyhmm, more likely 9V.
17:44:48speachy"optimal" is 10V apparently, so go with 9V.
17:45:01speachyhttps://www.rockbox.org/wiki/BatteryFAQ
17:47:53speachyQ11 in particular: 9V 600mA, center positive.
18:00
18:08:34 Quit ac_laptop (Ping timeout: 248 seconds)
18:19:47blebI wish it had info on the barrel dimensions
18:20:23blebthe pj-007 power jack page says 1.3 x 4.2 mm
18:20:48blebbut searching for 9v 600ma 4.2mm isn't turning up much
18:23:41 Join thanosengine [0] (~thanos@user/thanosengine)
18:35:32***Saving seen data "./dancer.seen"
18:41:34 Join cockroach [0] (~blattodea@user/cockroach)
19:00
19:19:43 Quit cockroach (Quit: leaving)
19:54:35 Join massiveH [0] (~massiveH@2600:4040:a982:5400:eddb:3355:1201:322d)
20:00
20:35:36***Saving seen data "./dancer.seen"
21:00
21:29:33 Quit toruvinn (Ping timeout: 252 seconds)
22:00
22:34:32 Quit Moriar (Ping timeout: 268 seconds)
22:35:40***Saving seen data "./dancer.seen"
23:00
23:21:36 Quit massiveH (Quit: Leaving)

Previous day | Next day