| 00:36:49 | _bilgus_ | well it works now but its kinda chunky and stutters a bit the current one does too at start but the inconsistent scroll speed in 570 makes it chunky jittery feeling |
| 00:40:54 | | Join jn_ [0] (~quassel@2a0a-a549-fc5e-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de) |
| 00:40:54 | | Quit jn_ (Changing host) |
| 00:40:54 | | Join jn_ [0] (~quassel@user/jn/x-3390946) |
| 00:41:06 | | Quit jn (Ping timeout: 248 seconds) |
| 00:43:02 | *** | Saving seen data "./dancer.seen" |
| 00:56:14 | | Quit spork (Quit: leaving) |
| 01:00 |
| 01:08:33 | | Join spork [0] (topic@77-164-207-8.fixed.kpn.net) |
| 02:00 |
| 02:03:58 | | Quit massiveH (Quit: Leaving) |
| 02:43:06 | *** | Saving seen data "./dancer.seen" |
| 02:58:34 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 02:59:26 | OlsroFR | The iFlash modded iPod Mini is still playing music; during 17 hours from now on stock OS. It dont want to die. The same iPod with the same battery and the same iFlash adapter could just last around 7 hours on Rockbox :( |
| 03:00 |
| 03:52:49 | user890104 | is it possible that rockbox has a more conservative battery discharge protection and stops earlier? |
| 03:53:24 | user890104 | i.e. can you measure the battery voltage after every test? |
| 04:00 |
| 04:01:50 | OlsroFR | user890104 mmm but why could I get more than 15 hours of battery life using a real CF card then on Rockbox ? |
| 04:02:03 | OlsroFR | Also, when it shut down, I cannot wake it up on Stock OS and play music during hours; it's just dead |
| 04:07:20 | | Quit jn_ (Ping timeout: 260 seconds) |
| 04:07:34 | | Join jn [0] (~quassel@user/jn/x-3390946) |
| 04:34:56 | user890104 | i saw something related to power management in the code recently, let me find it |
| 04:37:13 | user890104 | It's hard to find it on the phone, but it was related to core_sleep(), which was mostly not supported, with the exception of one or two CPU types |
| 04:37:54 | user890104 | and a comment that it will affecr battery life on other targets (where it's not supported) |
| 04:41:44 | OlsroFR | since a recent patch made by +speachy, IDE power management has been "repaired" for iFlash boards in the sense that the code is now correctly executed. It probably works fully on some iPods that have been fully reverse engineered in this sense. The function is ide_power_enable in power-ipod.c |
| 04:42:40 | OlsroFR | The issue is that it looks like that the devs that reverse engineered did not cared enough at finding all the register to fully shutdown the CF/IDE controller. And they didn't had to bother much with that, because the original HDD and normal CF cards are listening to ATA commands to enter in a very low power consumption state |
| 04:43:07 | *** | Saving seen data "./dancer.seen" |
| 04:43:49 | OlsroFR | The Stock OS is doing some wizardry to completely cut all power related to the storage device |
| 04:44:28 | OlsroFR | Right now my Stock OS iPod Mini is still playing music. Playing music since almost 19 hours non stop... :o |
| 04:45:00 | OlsroFR | It's almost 3x better than Rockbox. And it's still going (but I think it will shut down pretty. soon from now) |
| 04:46:01 | spork | be careful with claims that people 'do not care enough' ? |
| 04:47:01 | OlsroFR | They couldn't care about something that didn't exist. SD mods are a recent thing ;) |
| 04:47:41 | spork | if it was easy, it would have been done |
| 04:56:08 | OlsroFR | but since there is a small difference between Rockbox and Stock OS in terms of battery life even if you use a real CF card, I feel like this 2/3 hours difference of runtime using a real HDD or CF card may also be fixed if someone ever understand how to cut down power fully. I can remember getting 17/18 hours on Stock OS while getting 15 hours on |
| 04:56:09 | OlsroFR | Rockbox using a real CF card. Using ATA command forces the HDD/CF card to enter in a very low power consumption state, but this low power mode is still known to draw some power. |
| 04:57:15 | OlsroFR | So if this is figured out, I will also do benchmarks using this setup to confirm if there is improvements also for these standard setups. |
| 05:00 |
| 05:01:16 | user890104 | OlsroFR: a possible way to test it would be to make a custom build that turns off every pmu output one at a time, and then tries to access the storage, until it fails, if I had a Mini and an iFlash board, I'd be able to do it |
| 05:02:14 | user890104 | on the newer ipods it's memory-mapped, and clearing a bit or setting a register to zero volts would cut the power |
| 05:02:46 | user890104 | I believe it's the same on the older models, but never looked into PP-based ipods, as I don't have any |
| 05:02:53 | OlsroFR | If you can learn to me how to cut all pmu outputs, I would have the knowledge to do the rest and the trial and error myself even if it takes hours |
| 05:04:12 | user890104 | if you can compile a build of rockbox yourself, I can send you the changes needed to test this theory |
| 05:04:13 | OlsroFR | if (on) |
| 05:04:13 | OlsroFR | { |
| 05:04:14 | OlsroFR | GPIO_CLEAR_BITWISE(GPIOJ_OUTPUT_VAL, 0x04); |
| 05:04:14 | DBUG | Enqueued KICK OlsroFR |
| 05:04:14 | OlsroFR | DEV_EN |= DEV_IDE0; |
| 05:04:15 | OlsroFR | } |
| 05:04:15 | *** | Alert Mode level 1 |
| 05:04:15 | OlsroFR | else |
| 05:04:16 | *** | Alert Mode level 2 |
| 05:04:16 | OlsroFR | { |
| 05:04:16 | *** | Alert Mode level 3 |
| 05:04:16 | OlsroFR | DEV_EN &= ~DEV_IDE0; |
| 05:04:17 | *** | Alert Mode level 4 |
| 05:04:17 | OlsroFR | GPIO_SET_BITWISE(GPIOJ_OUTPUT_VAL, 0x04); |
| 05:04:17 | *** | Alert Mode level 5 |
| 05:04:17 | OlsroFR | } |
| 05:04:18 | *** | Alert Mode level 6 |
| 05:04:18 | OlsroFR | Here is the code that is called on the ipod mini to cut power |
| 05:04:27 | OlsroFR | Yes of course, I have all the tools to compile builds and do any kind of tests ;) |
| 05:04:56 | OlsroFR | I already submitted also several patches here :) |
| 05:05:23 | user890104 | I see, I'll look into this when I get home |
| 05:05:58 | user890104 | basucally there should be other registets like DEV_IDE0, but undocumented |
| 05:06:24 | user890104 | and reverse-ebgineering is one way to do it, experiments is the other |
| 05:07:04 | OlsroFR | thank you very much; I have plenty of time and motivation to work on this, but can't find the "frame of work" to canalize my energy to move on this subject by myself. Once I will get a frame of work, I can spend hours experimenting until finding the correct register that will improve battery life without breaking something other |
| 05:07:07 | user890104 | I wonder why it clears a GPIO output when powering on |
| 05:09:18 | user890104 | what I would do for such tests, in order to make them take less than a minute and not 15+ hours, is to desolder one of the battery cables, and attach a multimeter in sequence, set to Amps/miliamps. then after changing the DEV_EN registers, I would observe the change and take note |
| 05:12:03 | OlsroFR | I don't have that kind of hardware and skills. But I also feel like the test will not take 15 hours, only one or two hours |
| 05:12:15 | OlsroFR | I can notice that the battery percentage is decreasing much faster than it should |
| 05:12:33 | OlsroFR | The battery percentage is fortunately pretty precise on Rockbox for the iPod minis |
| 05:13:14 | OlsroFR | also much more precise than the Stock OS that can last 10 hours with "0%" of battery... On Rockbox when it decreases to 0, it's going to shut down soon |
| 05:14:19 | *** | Alert Mode OFF |
| 05:35:48 | | Quit PheralSparky (Quit: Leaving) |
| 06:00 |
| 06:13:53 | OlsroFR | the stock os iflash ipod is still running. 20H12 of continuous playback and still going strong. I would not have expected this to run so long. I can't remember how I did my first test months ago, but I maybe did it with ALAC which could explain the few hours less of battery |
| 06:14:47 | OlsroFR | the battery indicator still indicates the battery to be empty, but the music continues to play just fine |
| 06:15:40 | OlsroFR | https://support.apple.com/en-lb/112317 apple marketed the 2nd gen ipod mini to last 18 hours. So using a cameron sino battery + SD mod allows it to play music for longer than what apple themselves estimated with 128kbps AAC CBR files |
| 06:34:17 | user890104 | the original firmware might be draining thhe battery to unsafe levels, causing permanent damage to the cells, that would explain why it shows 0% for 10 hours |
| 06:35:12 | user890104 | you can do the same in rockbox by disabling these protections, and the PMU will probably be able to handle the voltage drop, bur that won't be sustainable in the long term |
| 06:43:08 | *** | Saving seen data "./dancer.seen" |
| 06:49:25 | OlsroFR | I don't feel like disabling the small reservoir will lead to 3x more battery life anyway |
| 06:51:50 | OlsroFR | the stock os firmware is just very imprecise also about how it report battery life |
| 06:52:02 | OlsroFR | there's just 3 or 4 visual levels, and you have to guess what does it mean really |
| 06:52:13 | OlsroFR | the empty battery icon may not mean "0%", it's just how interpret it |
| 06:52:24 | OlsroFR | how I interpret it* |
| 06:53:37 | OlsroFR | Rockbox also have issues with battery percentage, but not on this iPod. Look this benchmark on my iPod Video with a 3000mah battery + CF card, I could achieve 60 hours of battery life and it could last 12 hours of battery with 0%: https://pastebin.com/92D9S66X |
| 06:54:26 | OlsroFR | but on the Mini what Rockbox reports is very accurate, even when using a third party battery like the cameron sino ones |
| 07:00 |
| 07:25:59 | speachy | user890104: there's not really a "PMU" on the mini2g. There's a battery charger/regulator thingey but it only has a handful of ouputs and they're not switchable via GPIOs. |
| 07:27:27 | speachy | as I wrote earlier, the HDD is most likely hooked up to its own regulator connected directly to the battery (due to the inrush current necessary to spin up the drive) |
| 07:27:45 | speachy | and that regulator is probably switched via a GPIO-controlled MOSFET |
| 07:29:05 | speachy | the DEV_EN register likely clock-gates stuff internal to the SoC. It's that GPIO write that does the actual power supply killing |
| 07:29:22 | | Quit jn (Ping timeout: 252 seconds) |
| 07:29:45 | | Join jn [0] (~quassel@2001-4dd3-c765-0-20d-b9ff-fe49-15fc.ipv6dyn.netcologne.de) |
| 07:29:45 | | Quit jn (Changing host) |
| 07:29:45 | | Join jn [0] (~quassel@user/jn/x-3390946) |
| 07:30:07 | speachy | what's probably necessary here (given that the folks who did the original reverse engineering are long gone) is to disassemble the mini2g and map out the path from the disk connector to wherever its power comes from |
| 07:31:45 | speachy | and then figure out where the control line for that supply goes. that will probably require stripping components off the PCB due to hidden pads. |
| 07:32:50 | speachy | whatever it goes to, it's probably not GPIOJ2! |
| 07:34:35 | speachy | another approach would be to add a probe wire to the hard drive, write some custom debug code that toggles each GPIO bit and see what kills power. |
| 07:35:58 | speachy | of course there are other components with GPIO outputs (eg the battery management IC) that could also be responsible. so tracing the power feed as much as possible is still worth doing. |
| 07:37:24 | speachy | meanwhile, OlsroFR, the battery discharge curve (and thus the percentage reported) is different depending on the battery chemistry and other such things. Same with the emergency low power cutoff point. |
| 07:38:54 | speachy | so the curves in rockbox were measured against the original apple batteries, but 3rd party batteries are anyone's guess. You could generate a curve that's correct for what you have, but someone else buying the "same" battery a month from now could end up with sometihng completely different internally. |
| 07:49:57 | spork | the battery discharge curve is compiled into rockbox, right? |
| 07:50:14 | spork | no way to do you own benchmark and use that from then on? |
| 07:50:28 | spork | (without compiling a build using that curve) |
| 07:56:12 | speachy | _bilgus_: FS #13523 looks like a null pointer deref. I wonder if it would also happen if we try to work with something that's unaligned. |
| 07:56:13 | rb-bluebot | https://www.rockbox.org/tracker/task/13523 Daily builds since (early?) November crash at boot (bugs, unconfirmed) |
| 07:57:35 | speachy | my gut is telling me it has to do with the tag_table change, but I asked the reporter to narrow down which daily caused the problems. |
| 07:57:55 | speachy | spork: currently, yes. |
| 07:58:29 | spork | being able to do that could be welcome feature |
| 07:58:55 | spork | (patches welcome, i know) |
| 07:59:28 | speachy | yep |
| 08:00 |
| 08:07:36 | speachy | could easily just be a text file with 11 entries, one for each 10% mark. |
| 08:08:31 | speachy | maybe 12, with,the final being the low voltage cutoff point. The sfe value for that also depends on the battery chemistry. |
| 08:10:51 | speachy | sorry, 13. 11 for the main table, one for "dangerous" and one for "shutoff" |
| 08:12:37 | speachy | there's also a separate table for charging too. |
| 08:14:26 | spork | that was what i was thinking too. if a file is found, use that, otherwise the compiled-in default |
| 08:17:19 | OlsroFR | would be great to have a tool (PC ?) that can parse a battery benchmark once to auto-adjust (auto create a battery profile) from that file |
| 08:17:29 | OlsroFR | so creating it would be user friendly |
| 08:17:48 | spork | battery bench could just write those 13 values to a text file |
| 08:17:48 | speachy | or just have the battery benchmark plugin do it. |
| 08:17:51 | OlsroFR | yeah |
| 08:17:58 | spork | that is more user friendly than an external process |
| 08:18:13 | OlsroFR | well, speaking again of battery, the Stock OS testing iPod is still playing music. 22H15 of continous playback from now and still going. |
| 08:19:07 | OlsroFR | the mini 2nd gen is an awesome device |
| 08:19:44 | speachy | it's a good form factor let down by the limitations of the clickwheel. |
| 08:22:04 | OlsroFR | the clickwheel is very pleasant to use in my opinion, even more than a 4th gen mono that tends to "freeze" and that need to lock then unlock again the ipod to fix it |
| 08:22:23 | OlsroFR | on the mini it is very reliable. The main limitation in my opinion is the screen, especially on Stock OS; on Stock OS, the text can't scroll in menus |
| 08:23:01 | OlsroFR | Rockbox can use the limited screen space much more efficiently and navigation through the database is very fast if everything is correctly tagged |
| 08:25:58 | OlsroFR | the main limitation to me comes around repairability. It seems almost impossible to open one in the first time without doing some slight visual damages considering how glued things are. Also, the pins from the bottom plastic cover are very fragile and considering how they need to be perfected with the metal surface just below, it's also pretty easy |
| 08:25:58 | OlsroFR | to damage it enough to require super-glue or buy a replacement for this specific part. |
| 08:25:59 | OlsroFR | Ironically, I now find the classics and the videos easier to open now that I do it with the iSesamo compared to a Mini the first time |
| 08:28:21 | OlsroFR | the 4th gen mono is my preferred in terms of repairability from the iPods that I own (minis, video, classics, and that 4th gen mono). It has the old battery connector which is not fragile, and the plastic shell goes out so easily using an iSesamo. It seems very difficult to damage it while opening. When I open a mini or classic/video, I go very |
| 08:28:21 | OlsroFR | slowly |
| 08:35:30 | _bilgus_ | speachy FS crash ack, will wait for follow up before I start screwing around w/ it |
| 08:38:40 | speachy | I think I'm going to take a chance on this: https://www.ebay.com/itm/126790924701 |
| 08:38:44 | _bilgus_ | user890104, re the current measuring, I did similar with the Clip+ but your digital multimeter won't work all that well, an analog meter works decently as long as the range is small enough ultimately a precision shunt resistor ( has to be pretty small R value) and a scope |
| 08:40:03 | _bilgus_ | it doesn't have to be a real good scope I did it with a 80s era techtronix 1MHZ IIRC |
| 08:42:24 | _bilgus_ | and if you are just actively doing things the analog meter is the way to go, it auto averages for you and if the nedel is lower you are saving power |
| 08:42:31 | _bilgus_ | needle* |
| 08:42:54 | _bilgus_ | you just can't quantify 'How Much' |
| 08:43:12 | *** | Saving seen data "./dancer.seen" |
| 08:44:56 | speachy | and this: https://www.ebay.com/itm/156107451888 |
| 08:44:59 | _bilgus_ | 20GB and a 1 year warranty! |
| 08:45:13 | speachy | $60 for an iRiver remote.. ugh. |
| 08:45:36 | _bilgus_ | I know some people who'd have killed for that back in the day |
| 08:45:59 | speachy | well the "for parts" H320 is now mine, I guess I'll find out what doesn't work. battery and/or hard drive, most likely. |
| 08:46:05 | _bilgus_ | note its more than the device it controls (as it should be :P) |
| 08:46:41 | speachy | "working" iHPs go for a lot more than $60. |
| 08:47:00 | _bilgus_ | heh |
| 08:48:50 | speachy | (pretty sure I have a spare HDD from an old ipod photo lying around, or I can slap a CF card in there instead) |
| 08:49:10 | speachy | (and it still uss polarity-reduced ipod batteries, so those should be easy enough to find still) |
| 08:50:53 | speachy | ok, snagged the remote too. figure _somoene_ here should have the means to test out the "remote lcd" portion of our codebase.. |
| 08:50:57 | OlsroFR | CF cards are very easy to find for very cheap in the second hand market, as long as you can satisfy yourself with 64GB or less |
| 08:51:58 | OlsroFR | Ali Express used to sell 256GB Kingspecs CF cards for pretty cheap (around 50 dollars), but these are now out of stock since weeks. These cards works perfectly with iPods |
| 08:52:06 | speachy | I don't intend to use this for anything other than basic testing/bugfixes. it's a bit bulky for my liking, plus it's waaaay underpowered even compared to the ipods. |
| 08:52:56 | OlsroFR | haha even iPods have the brain of an ant compared to modern standards ;( |
| 08:53:32 | OlsroFR | I never saw a video of how good Rockbox could run on modern DAPs like Surfans F20/HiFi Walker. I am wondering if it feels like a different experience in terms of responsiveness with rockbox |
| 08:54:10 | spork | responsiveness in what area ? |
| 08:55:18 | OlsroFR | UI, songs loading, boot time... |
| 08:55:32 | OlsroFR | My ipod need 15 seconds to boot then cache the database in RAM |
| 08:55:46 | OlsroFR | (around 15 seconds, did not benchmarked it exactly) |
| 08:57:05 | spork | i have never used the database on a quick device |
| 08:57:15 | spork | so could not tell you how that goes |
| 08:57:49 | speachy | the raw CPU has plenty of oomph, but it's hamstrung by a lack of data cache (really hurting memory performance) |
| 09:00 |
| 09:00:02 | | Quit JanC (Remote host closed the connection) |
| 09:00:15 | spork | x1000e seems fast but what i do on a clip+ also works fast enough. i just used it differently. |
| 09:01:52 | speachy | x1000 has a modern MIPS core clocked at 1GHz. it's by far the most performant CPU core we support |
| 09:03:02 | speachy | (second fastest is the jz4760, 600MHz MIPS, but it's let down by a few quirks, ntably not being able to do double-buffered DMA) |
| 09:03:23 | speachy | we don't have _any_ codec optimizatons for the mips targets |
| 09:03:31 | speachy | and it doesn't matter |
| 09:04:07 | | Join JanC [0] (~janc@user/janc) |
| 09:05:28 | spork | only thing from the OF of both my x1000 devices is the it can idle without shutting down |
| 09:06:18 | spork | speachy, how much of a 'hosted' port is hardware specific ? |
| 09:06:45 | spork | i imagine things like lcd, input and perhaps audio ? |
| 09:08:05 | speachy | tehy all use standard linux userspace APIs but the specifics matter −− eg lcd resolution/color depth, scancode -> button mappings (and how many input devices need to be monitored), and the name of the audio device and mixer controls. |
| 09:08:42 | speachy | also battery/power/platform management varies (eg how to read the battery level, charging status, reboot or shut down the device) |
| 09:08:43 | spork | no so much hardcore hardware stuff then |
| 09:09:31 | spork | the hiby r1 has all its drivers as modules, they decompile really well in ghidra |
| 09:09:36 | speachy | and the platform integration varies widely too −− location of storage and other filesystem stuffs, installation method, etc. |
| 09:10:20 | speachy | I'm rather surprised by that, but good for them for finally trying to do things "right" |
| 09:10:40 | | Join Skyrider [0] (~Skyrider@static.219.32.201.138.clients.your-server.de) |
| 09:10:44 | spork | it is also linux 4.x instead of the m3k/q1 and perhaps aigo |
| 09:10:55 | spork | it looks quite clean and organized |
| 09:11:21 | speachy | the m3k SW wasn't done by hiby, unlike the remaining X1000 targets we support |
| 09:12:06 | spork | shanling pretends to do their own software |
| 09:13:17 | speachy | I tkae that back, the Q1 wasn't hiby-powered either |
| 09:14:24 | speachy | all of the hibyos players (so far) work the same way. |
| 09:15:05 | spork | which would be a good hosted port to study for inspiration ? |
| 09:15:06 | speachy | trivial to patch our stuff into them, and porting rockbox is just a matter of dealing with all that device-specific glue code. |
| 09:15:42 | speachy | I'd say go for the agptek rocker. it's probably the simplest of the lot |
| 09:15:52 | spork | you are welcome to trivially patch the r1's update file to allow running stuff from the sd card (like rockbox eventually) |
| 09:16:26 | spork | it is a nice piece of fairly cheap hardware that is held back by the software |
| 09:16:33 | spork | and it is new and readily available |
| 09:16:37 | speachy | to even generate the "bootloader" the display and input stuff needs to be ported. |
| 09:17:04 | spork | i was wondering about that |
| 09:17:14 | speachy | step 1, patch in adb so you can go spelunking on the live system |
| 09:17:26 | spork | adb is there but not enabled in the kernel so not active |
| 09:17:31 | speachy | step 2, hack the codebase enough to build a bootloader |
| 09:17:36 | speachy | step 3, everything else. |
| 09:17:50 | spork | i know their player is started from an init script |
| 09:17:53 | speachy | well, step 3 is "everything other than the plugins" |
| 09:18:01 | spork | why do you not run rockbox from there ? |
| 09:18:02 | speachy | step 4 is "plugins" which is mostly keymap crap. |
| 09:18:25 | spork | check if some file exists on the sd card, if so run rockbox |
| 09:18:49 | speachy | the bootloader (really, just a boot _menu_) allows users to choose which player is launched, control debugging options, etc. |
| 09:19:01 | spork | sure, but that is hard :) |
| 09:19:26 | speachy | no, that's _easy_ because it only uses a small subset of the functionality that rockbox needs. |
| 09:19:36 | | Quit IPG (Read error: Connection reset by peer) |
| 09:19:43 | spork | in the shanling q1 fireware there was a check in some init file for a weird file name that could let you execute your own stuff |
| 09:19:48 | spork | firmware |
| 09:20:20 | speachy | and the hibyos stuff has an inits cript that just launches hiby_player as the final invocation |
| 09:20:25 | spork | yes |
| 09:20:49 | speachy | we patch ourselves in there by renaming that binary in favor of the bootmenu thing. |
| 09:21:04 | spork | aha |
| 09:21:05 | speachy | ie mininal changes |
| 09:21:44 | speachy | especially since the stock player has functionality we don't −− mainly bluetooth. |
| 09:22:02 | speachy | hmm, I wonder if the R1 still uses bluez4. |
| 09:22:28 | speachy | they updated the krnel considerably, so maybe they also updated the rest of userspace to be less than a decade old? |
| 09:24:04 | spork | bluez5 i think |
| 09:24:07 | spork | let me check |
| 09:27:28 | speachy | if they updated their userspace we might have to put together a difeferent cross-compiler for it. |
| 09:27:46 | speachy | (the actual compiler probably doesn't matter but the set of libraries we have to build around it does) |
| 09:28:43 | spork | Linux version 4.4.94+ (fkp@hibyserver) (gcc version 5.2.0 (Ingenic r3.2.1-gcc520 2017.12-15) ) #623 PREEMPT Mon Nov 18 16:43:27 CST 2024 |
| 09:30:40 | spork | i can only find bluez test scripts |
| 09:31:15 | speachy | most of the extra crap we build is nonl needed for bluetooth support. |
| 09:34:54 | spork | i do not know why i said i think it is bluez5, i cannot find any trace of it now |
| 09:37:02 | | Quit JanC (Remote host closed the connection) |
| 09:39:08 | | Join JanC [0] (~janc@user/janc) |
| 09:45:02 | | Quit JanC (Remote host closed the connection) |
| 09:45:58 | | Join JanC [0] (~janc@user/janc) |
| 09:54:37 | spork | i guess i need to figure out how to trivially patch things :) |
| 10:00 |
| 10:06:10 | OlsroFR | Ok so my iPod Mini is running on perpetual power or something :D This iFlash modded on Stock OS is still running and playing music on the same single charge since 24 hours from now: a complete day. It's going to last 4x more than Rockbox, can't believe it, I would have expected it to shut down way sooner than that |
| 10:16:42 | _bilgus_ | then it sounds like a 3x improvement should be possible if you can figure out the why |
| 10:24:45 | _bilgus_ | I guess you could compare to a run against your harddrive or a cf card but ultimately how would you figure out if it was actually the failure (misconfiguration, lack of gpio control, etc) or some other portion of the system.. |
| 10:26:31 | _bilgus_ | for instance maybe say we had a bad power management situation that kept clocks running fast or a dcdc setting that burns power, maybe with the hardrive eating power it is not noticed so much |
| 10:30:11 | _bilgus_ | I guess you could do a run of a very short track on repeat (so it stays in ram) and compare runtimes then |
| 10:34:35 | _bilgus_ | if its the drive then the OF and rockbox should pretty much match right? but if its 20% higher that would be a huge drain (just not compared to a harddrive) |
| 10:37:52 | | Quit OlsroFR (Quit: Client closed) |
| 10:43:16 | *** | Saving seen data "./dancer.seen" |
| 10:44:39 | | Join retr0id2 [0] (~Retr0id@user/retr0id) |
| 10:44:55 | | Join amiconn_ [0] (jens@p200300ea8711db00305e95fffec66ff3.dip0.t-ipconnect.de) |
| 10:44:55 | | Nick amiconn is now known as Guest6320 (jens@p4fe76649.dip0.t-ipconnect.de) |
| 10:44:55 | | Quit Guest6320 (Killed (tungsten.libera.chat (Nickname regained by services))) |
| 10:44:55 | | Nick amiconn_ is now known as amiconn (jens@p200300ea8711db00305e95fffec66ff3.dip0.t-ipconnect.de) |
| 10:45:09 | | Quit dbohdan (Quit: ZNC 1.8.2+deb3.1 - https://znc.in) |
| 10:45:28 | | Join dbohdan [0] (~dbohdan@user/dbohdan) |
| 10:45:37 | | Quit Arsen (Remote host closed the connection) |
| 10:46:05 | | Quit retr0id (Read error: Connection reset by peer) |
| 10:46:05 | | Nick retr0id2 is now known as retr0id (~Retr0id@user/retr0id) |
| 10:46:42 | | Quit aaabbb (Ping timeout: 252 seconds) |
| 10:46:47 | | Join Arsen [0] (arsen@gentoo/developer/managarm.dev.Arsen) |
| 10:47:27 | | Join aaabbb [0] (sitku@bitcoinshell.mooo.com) |
| 10:52:28 | speachy | _bilgus_: that's why I want to see a measurement on the VCC rail of the IDE connector |
| 10:53:25 | speachy | I'm willing to bet that isn't being turned off, which means the SD adapter and the SD cards themselves remain powered. |
| 11:00 |
| 11:05:28 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 11:07:38 | OlsroFR | _bilgus Good idea. But this would imply to open, do swaps, do benchmarks, etc. Maybe later. Also, in terms of personal interests, I will not be motivated to search for more if it is just about 20%. I was really satisfied from my blue iPod Mini 2nd gen who can last between 15 to 16 hours using a real CF card. Getting at least 15 hears would be |
| 11:07:39 | OlsroFR | enough to satisfy me and consider it as a reliable device that will always have some juice when I will take it. But even if Rockbox brings so much compared to the Stock OS, having just 6 hours of battery life is way too low to give me satisfaction |
| 11:08:02 | OlsroFR | Also, by the way, the iFlash modded Stock OS iPod Mini 2nd gen is still not dead. Still running strong; 25 hours of runtime. It definitely don't want to die |
| 11:09:30 | OlsroFR | I am wondering if it will last 30 hours, which would make it last just like Apple marketed their last iPod Classics... Also it is kinda cool to have much more battery runtime than modern tech devices with soldered batteries like Surfans, Hybi R1, etc with something that is that old |
| 11:10:55 | OlsroFR | The reference of my battery on that mini is the CS-EC003XL. It is marketed to be an "high capacity battery" for the ipod mini. The mah value is maybe wrong, who knows (but I don't think cameron sino is lying on its batteries), but I do know that original iPod Mini batteries have much less capacity than that |
| 11:40:33 | | Join JanC_ [0] (~janc@user/janc) |
| 11:40:34 | | Quit JanC (Killed (zirconium.libera.chat (Nickname regained by services))) |
| 11:40:34 | | Nick JanC_ is now known as JanC (~janc@user/janc) |
| 12:00 |
| 12:08:26 | OlsroFR | still running; 26 hours of total runtime |
| 12:12:52 | OlsroFR | it just stopped ! |
| 12:16:37 | OlsroFR | by the way, this ipod has a samsung evo microSD (512GB) in a sandisk SD adapter |
| 12:43:17 | *** | Saving seen data "./dancer.seen" |
| 12:45:44 | | Quit OlsroFR (Quit: Client closed) |
| 13:00 |
| 13:16:20 | speachy | would be interesting to compare the battery life of the ipod4g (either the greyscale or "photo" version) vs the mini2g. |
| 13:17:47 | speachy | as they are essentially the same hardware. |
| 13:37:16 | | Quit buZz (Read error: Connection reset by peer) |
| 13:37:35 | | Join buZz [0] (~buzz@192.161.48.59) |
| 14:00 |
| 14:17:15 | | Join PheralSparky [0] (~S|h|a|w|n@user/shawn/x-4432647) |
| 14:43:19 | *** | Saving seen data "./dancer.seen" |
| 15:00 |
| 15:14:59 | rb-bluebot | Build Server message: New build round started. Revision afb0e845cf, 345 builds, 9 clients. |
| 15:15:00 | rb-bluebot | strcasestr smaller function for PREFER_SIZE_OVER_SPEED by William Wilgus |
| 15:19:42 | | Join lebellium [0] (~lebellium@2a01cb0405d07f00adae87b9bb1a46e0.ipv6.abo.wanadoo.fr) |
| 15:27:33 | speachy | w |
| 15:30:53 | rb-bluebot | Build Server message: Build round completed after 954 seconds. |
| 15:30:55 | rb-bluebot | Build Server message: Revision afb0e845cf result: All green |
| 15:34:46 | speachy | hmm, do we really want PREFER_SIZE_OVER_SPEED for most targets? |
| 15:35:48 | speachy | I guess it's picked up automatically by __OPTIMIZE_SIZE__ that we use globally |
| 15:38:50 | _bilgus_ | I'd argue no for RAM > 8 mb |
| 15:40:24 | _bilgus_ | possibly no for strlen too |
| 15:40:30 | _bilgus_ | it makes a big difference |
| 15:49:36 | speachy | that crash on iriver started with the 20241117 nightly, supposedly |
| 16:00 |
| 16:00:43 | | Quit speachy (Quit: WeeChat 4.4.3) |
| 16:09:37 | | Join chris_s [0] (~chris_s@2a09:bac3:2d1e:2478::3a2:34) |
| 16:10:50 | chris_s | huh... splashf goes boom in the simulator, too... |
| 16:11:09 | chris_s | (h300) |
| 16:15:37 | | Quit Arsen (Quit: No Ping reply in 180 seconds.) |
| 16:17:02 | | Join Arsen [0] (arsen@gentoo/developer/managarm.dev.Arsen) |
| 16:27:44 | rb-bluebot | Build Server message: New build round started. Revision b1e70db6e7, 345 builds, 9 clients. |
| 16:27:45 | rb-bluebot | Fix FS #13523 splashf crash on devices with remote LCD by Christian Soffke |
| 16:32:59 | | Join Everything [0] (~Everythin@46-133-159-79.mobile.vf-ua.net) |
| 16:42:23 | rb-bluebot | Build Server message: Build round completed after 880 seconds. |
| 16:42:25 | rb-bluebot | Build Server message: Revision b1e70db6e7 result: All green |
| 16:43:23 | *** | Saving seen data "./dancer.seen" |
| 16:55:24 | chris_s | it may make sense to release a new iPod color/photo bootloader (although for some reason, it only boots successfully for me when built using the newer toolchain - GCC 9.5.0 - instead of GCC 4.9.4). |
| 16:55:27 | chris_s | At least in my experience, the released version randomly throws.a "bad checksum" error every few boots. Completely stock hardware and I don't *think* the drive is to blame. Error doesn't happen anymore using a bootloader built from the latest code. On the other hand, I haven't really seen any other reports about this |
| 16:56:31 | chris_s | Don't know if anyone here can reproduce that behavior |
| 17:00 |
| 17:13:43 | chris_s | @OlsroFR re what you wrote about the 4th gen mono clickwheel "freezing", requiring you to lock and unlock it again? I haven't experienced anything like that. Does it happen only using Rockbox or using the OF as well? Is it possible your clickwheel is defective? |
| 17:17:42 | | Quit lebellium (Quit: Leaving) |
| 17:40:13 | | Join speachy [0] (~speachy@rockbox/developer/speachy) |
| 17:40:13 | Mode | "#rockbox +v speachy" by ChanServ (ChanServ@services.libera.chat) |
| 17:41:01 | speachy | chris_s: You mean to say the current codebase doesn't build a clean bootloader with the 4.9 toolchain but does with 9.5? |
| 17:41:39 | speachy | but modern code is a _lot_ more robust than the last released bootloader build; especially with non-stock drives. |
| 17:42:39 | speachy | thanks for sorting out that ihp crash, I had been assuming it was disk or enum-related but didn't have a chance to look deeper when the reporter narrowed down when it broke |
| 17:43:21 | chris_s | re the bootloader: at least in the case of the iPod photo that has been my somewhat surprising experience, yes. |
| 17:43:49 | chris_s | although maybe I should triple-check I didn't make a mistake somewhere |
| 17:44:05 | speachy | That warrants some spelunking into the generated binary, assuming no mistakes on your part |
| 17:54:14 | | Quit Everything (Quit: leaving) |
| 18:00 |
| 18:02:37 | | Quit JanC (Remote host closed the connection) |
| 18:02:41 | | Join OlsroFR [0] (~OlsroFR@user/OlsroFR) |
| 18:03:36 | OlsroFR | chris_s This happen in general when I don't use it for hours. In normal usage, it never "soft lock" suddently and works like a charm. Once I do the "lock then unlock" trick, it directly restart working just fine during the whole music listening session. |
| 18:03:57 | | Join JanC [0] (~janc@user/janc) |
| 18:04:27 | OlsroFR | I had only one ipod 4th gen mono of my life and it is not physically in perfect shape, but it is very reliable. Can't tell if the clickwheel itself is defective, but I know that this issue happen in both Stock OS or Rockbox from my memories |
| 18:05:03 | OlsroFR | but I didn't use much this iPod with Stock OS unfortunately so I am not 100% sure about this |
| 18:05:05 | | Quit OlsroFR (Client Quit) |
| 18:08:18 | | Join JanC_ [0] (~janc@user/janc) |
| 18:08:19 | | Quit JanC (Killed (zirconium.libera.chat (Nickname regained by services))) |
| 18:08:19 | | Nick JanC_ is now known as JanC (~janc@user/janc) |
| 18:13:31 | | Nick JanC is now known as Guest3827 (~janc@user/janc) |
| 18:13:31 | | Join JanC_ [0] (~janc@user/janc) |
| 18:13:31 | | Quit Guest3827 (Killed (molybdenum.libera.chat (Nickname regained by services))) |
| 18:13:31 | | Nick JanC_ is now known as JanC (~janc@user/janc) |
| 18:43:27 | *** | Saving seen data "./dancer.seen" |
| 19:00 |
| 19:02:27 | | Quit chris_s (Quit: Client closed) |
| 19:22:10 | | Quit q3k (Ping timeout: 248 seconds) |
| 19:24:18 | | Join q3k [0] (q3k@boston-packets.hackerspace.pl) |
| 19:24:59 | speachy | heads-up, server is going down for an upgrade. shouldn't be more than 15 minutes. |
| 19:25:04 | DEBUG | Received signal 15 (SIGTERM), terminating (snapshot: transfer.c line 304) |
| 19:25:04 | *** | Cleanup |
| 19:25:04 | *** | Saving seen data "./dancer.seen" |
| 19:25:04 | *** | Exit |
| 19:48:54 | *** | Started Dancer V4.16 |
| 19:48:54 | DEBUG | gethostbyname(2) failed for irc.libera.chat (Resource temporarily unavailable) (snapshot: netstuff.c line 99) |
| 19:48:54 | DEBUG | gethostbyname(2) failed for irc.libera.chat (Resource temporarily unavailable) (snapshot: netstuff.c line 99) |
| 19:48:54 | *** | Connected to irc.libera.chat on port 6667 |
| 19:48:54 | *** | Logfile for #rockbox started |
| 19:48:58 | Mode | "rb-logbot :+iw" by rb-logbot |
| 19:49:04 | *** | Server message 501: 'rb-logbot :Unknown MODE flag' |
| 19:49:05 | | Join rb-logbot [0] (~rockbox@archos.rockbox.org) |
| 19:49:05 | | Join q3k [0] (q3k@boston-packets.hackerspace.pl) |
| 19:49:05 | | Join JanC [0] (~janc@user/janc) |
| 19:49:05 | | Join +speachy [0] (~speachy@rockbox/developer/speachy) |
| 19:49:05 | | Join Arsen [0] (arsen@gentoo/developer/managarm.dev.Arsen) |
| 19:49:05 | | Join PheralSparky [0] (~S|h|a|w|n@user/shawn/x-4432647) |
| 19:49:05 | | Join buZz [0] (~buzz@192.161.48.59) |
| 19:49:05 | | Join aaabbb [0] (sitku@bitcoinshell.mooo.com) |
| 19:49:05 | | Join dbohdan [0] (~dbohdan@user/dbohdan) |
| 19:49:05 | | Join amiconn [0] (jens@p200300ea8711db00305e95fffec66ff3.dip0.t-ipconnect.de) |
| 19:49:05 | | Join retr0id [0] (~Retr0id@user/retr0id) |
| 19:49:05 | | Join Skyrider [0] (~Skyrider@static.219.32.201.138.clients.your-server.de) |
| 19:49:05 | | Join jn [0] (~quassel@user/jn/x-3390946) |
| 19:49:05 | | Join spork [0] (topic@77-164-207-8.fixed.kpn.net) |
| 19:49:05 | | Join Nezumi-sama [0] (~narf@syn-067-053-148-069.biz.spectrum.com) |
| 19:49:05 | | Join LjL [0] (~ljl@user/ljl) |
| 19:49:05 | | Join paulk [0] (~paulk@about/aquilenet/user/paulk) |
| 19:49:05 | | Join mrkrisprolls [0] (~mrkrispro@lecturify.net) |
| 19:49:05 | | Join lunchd [0] (~launchd@bitbot/husky/launchd) |
| 19:49:05 | | Join braewoods__ [0] (~braewoods@user/braewoods) |
| 19:49:05 | | Join tobbez [0] (tobbez@user/meow/tobbez) |
| 19:49:05 | | Join martylake [0] (~martylake@82.66.174.242) |
| 19:49:05 | | Join Galois [0] (djao@efnet.math.uwaterloo.ca) |
| 19:49:05 | | Join pixelma [0] (marianne@p4fe76649.dip0.t-ipconnect.de) |
| 19:49:05 | | Join drew [0] (~drew@user/drew) |
| 19:49:05 | | Join jackie [0] (~jackie@banana-new.kilobyte22.de) |
| 19:49:05 | | Join cb [0] (~calvin@138.197.175.25) |
| 19:49:05 | | Join edhelas_ [0] (9d94237298@v2202101139504140605.quicksrv.de) |
| 19:49:05 | | Join jj5 [0] (~jj5@100.80.216.139.dynamic.dsl.dv.iprimus.net.au) |
| 19:49:05 | | Join jssfr [0] (~jssfr@95.217.50.17) |
| 19:49:05 | | Join baltazar [0] (~baltazar@user/baltazar) |
| 19:49:05 | | Join michaelni [0] (~michael@41.66.67.113) |
| 19:49:05 | | Join __builtin [0] (~quassel@rockbox/developer/builtin) |
| 19:49:05 | | Join ursa-major [0] (114efe6c39@2a03:6000:1812:100::11f3) |
| 19:49:05 | | Join dook [0] (c9db4a6425@2a03:6000:1812:100::1222) |
| 19:49:05 | | Join lane [0] (809450f172@2a03:6000:1812:100::1300) |
| 19:49:05 | | Join markun [0] (~markun@213-10-76-13.fixed.kpn.net) |
| 19:49:05 | | Join melmothX [0] (~marco@amusewiki/marco) |
| 19:49:05 | | Join bpye [0] (~bpye@user/bpye) |
| 19:49:05 | | Join jjs0077018310196 [0] (~jjs007@host86-172-82-147.range86-172.btcentralplus.com) |
| 19:49:05 | | Join Trzyzet [0] (~Trzyzet@88.97.248.229) |
| 19:49:05 | | Join aigo [0] (uid678396@user/Aigo) |
| 19:49:05 | | Join rnkn [0] (~rnkn@starbeastie.blip.town) |
| 19:49:05 | | Join akaWolf [0] (~akaWolf@akawolf.org) |
| 19:49:05 | | Join advcomp2019 [0] (~advcomp20@user/advcomp2019) |
| 19:49:05 | | Join hook54321 [0] (sid149355@user/hook54321) |
| 19:49:05 | | Join Natch [0] (~natch@c-92-34-7-158.bbcust.telenor.se) |
| 19:49:05 | | Join Bobathan_ [0] (~admin@syn-065-029-248-157.res.spectrum.com) |
| 19:49:05 | | Join cstine [0] (~cstine@150.136.136.191) |
| 19:49:05 | | Join TorC [0] (~Tor@fsf/member/TorC) |
| 19:49:05 | | Join rogeliodh [0] (~rogeliodh@rogeliodh.dev) |
| 19:49:05 | | Join npmania [0] (~npmania@210.123.73.190) |
| 19:49:05 | | Join Tonux [0] (~Tonux@193.32.127.240) |
| 19:49:05 | | Join vup [0] (~~~~@46.101.193.235) |
| 19:49:05 | | Join gevaerts [0] (~fg@user/gevaerts) |
| 19:49:05 | | Join _bilgus_ [0] (~bilgus@syn-162-154-213-134.res.spectrum.com) |
| 19:49:05 | | Join sch [0] (a2fe4b5ecb@irc.cheogram.com) |
| 19:49:05 | | Join Bubblegumdrop [0] (~znc@user/Bubblegumdrop) |
| 19:49:05 | | Join bluebrother [0] (~dom@user/bluebrother) |
| 19:49:05 | | Join sebagala [0] (~Burak@185.25.123.34) |
| 19:49:05 | | Join CH23 [0] (CH23@revspace/participant/ch23) |
| 19:49:05 | | Join b0 [0] (~b0@user/b0) |
| 19:49:05 | | Join GeekShadow [0] (~antoine@lbf.turmel.info) |
| 19:49:05 | | Join Malinux [0] (~malin@2001:4641:4dfa::12c:c4a7) |
| 19:49:05 | | Join troglodito [0] (~cave@81.4.123.134) |
| 19:49:05 | | Join Maxdamantus [0] (~Maxdamant@user/maxdamantus) |
| 19:49:05 | | Join cnx [0] (~cnx@tem.loang.net) |
| 19:49:05 | | Join acidsys [0] (~crameleon@openSUSE/member/crameleon) |
| 19:49:05 | | Join rasher [0] (~rasher@user/rasher) |
| 19:49:05 | | Join munkis [0] (~mendel_mu@2600:4041:5ac9:a100:b225:aaff:fe5e:10eb) |
| 19:49:05 | | Join prg [0] (~prg@xannode.com) |
| 19:49:05 | | Join ats [0] (~ats@cartman.offog.org) |
| 19:49:05 | | Join rudi_s [0] (~simon@user/rudi-s/x-7673890) |
| 19:49:05 | | Join ParkerR [0] (znc@znc.withg.org) |
| 19:49:05 | | Join SammysHP [0] (~SammysHP@faol.sammyshp.de) |
| 19:49:05 | | Join bbbccc [0] (~aaabbb@5.254.86.4) |
| 19:49:05 | | Join danwellby [0] (~danwellby@cynthiajndevey.plus.com) |
| 19:49:05 | | Join _alice [0] (sid544964@id-544964.ilkley.irccloud.com) |
| 19:49:05 | | Join Xeha [0] (~Xeha@user/Xeha) |
| 19:49:05 | | Join benjaoming [0] (~benjaomin@37.139.19.237) |
| 19:49:05 | | Join wLLm [0] (~wLLm@2a10:3781:3ed0::3) |
| 19:49:05 | | Join tertu [0] (~tertu@user/tertu) |
| 19:49:05 | | Join kirvesAxe [0] (kirvesaxe@user/kirvesaxe) |
| 19:49:05 | | Join yosafbridge [0] (~yosafbrid@static.38.6.217.95.clients.your-server.de) |
| 19:49:05 | | Join Ckat [0] (~Ckat@xn--z7x.xn--6frz82g) |
| 19:49:05 | | Join bertrik [0] (~bertrik@revspace/participant/bertrik) |
| 19:49:05 | | Join zou [0] (~zou@user/zou) |
| 19:49:05 | | Join olspookishmagus [0] (~pookie@83.212.124.75) |
| 19:49:05 | | Join bleb [0] (~cm@user/bleb) |
| 19:49:05 | | Join Ripcord [0] (xyz@66.254.101.193) |
| 19:49:05 | | Join toruvinn [0] (~toruvinn@KD119106027116.ppp-bb.dion.ne.jp) |
| 19:49:05 | | Join user890104 [0] (~Venci@freemyipod/user890104) |
| 19:49:05 | | Join rb-bluebot [0] (~rb-bluebo@rockbox/bot/utility) |
| 19:49:05 | | Join funman [0] (~fun@chui-pas.net) |
| 19:49:19 | speachy | 24, apparently. |
| 20:00 |
| 20:10:45 | | Quit npmania (Read error: Connection reset by peer) |
| 20:21:08 | | Join npmania [0] (~npmania@210.123.73.190) |
| 20:30:22 | | Join massiveH [0] (~massiveH@2600:4040:a982:5400:accc:53c2:79d4:38d6) |
| 20:58:52 | | Join COMPL_EXE [0] (~compl.exe@aosc/dev/origincode) |
| 21:00 |
| 21:48:56 | *** | Saving seen data "./dancer.seen" |
| 22:00 |
| 22:40:37 | | Quit COMPL_EXE (Read error: Connection reset by peer) |
| 22:40:59 | | Join COMPL_EXE [0] (~compl.exe@aosc/dev/origincode) |
| 23:00 |
| 23:08:41 | | Quit massiveH (Quit: Leaving) |
| 23:48:58 | *** | Saving seen data "./dancer.seen" |
icon identifies that the person is a core developer (has commit access).