| 00:35.31 | *** join/#maemo qwazix (~qwazix@Maemo/community/ex-council/qwazix) |
| 01:21.17 | *** join/#maemo pkill9 (~pkill9@unaffiliated/pkill9) |
| 01:27.00 | *** join/#maemo Natch (~Natch@h-155-122.A444.priv.bahnhof.se) |
| 01:29.47 | *** join/#maemo Tony` (~user@74-212-219-34.static-ip.telepacific.net) |
| 01:31.12 | *** part/#maemo Tony` (~user@74-212-219-34.static-ip.telepacific.net) |
| 15:56.01 | *** join/#maemo infobot (~infobot@rikers.org) |
| 15:56.01 | *** topic/#maemo is Welcome to #maemo http://maemo.org/intro/ | N9/harmattan related questions please in #harmattan ! | "firmware" see ~lf (or ~tabletsdev ~combined ~emmc ~flasher) sdk: ~sb | http://wiki.maemo.org/Community_Council | Source: http://wiki.maemo.org/Sources | Chanlog: http://mg.pov.lt/maemo-irclog | #maemo-ssu is where the (few) devels and maintainers of the maemo-future meet |
| 16:37.03 | *** join/#maemo Pali (~pali@Maemo/community/contributor/Pali) |
| 17:01.49 | *** join/#maemo louis_ (~louisdk@5.103.138.205) |
| 17:04.23 | *** join/#maemo pagurus (~user@i59F7C62E.versanet.de) |
| 17:22.14 | *** join/#maemo louis_ (~louisdk@5.103.138.205) |
| 17:40.18 | *** join/#maemo pkill9 (~pkill9@unaffiliated/pkill9) |
| 17:44.43 | *** join/#maemo Kabouik- (~kabouik@236.34.200.37.customer.cdi.no) |
| 17:53.45 | *** join/#maemo florian_kc (~florian_k@Maemo/community/contributor/florian) |
| 18:42.36 | *** join/#maemo eMHa (~mh@p200300738E05E700267703FFFE6CC360.dip0.t-ipconnect.de) |
| 19:05.34 | *** join/#maemo sicelo (~sicelo@Maemo/community/council/sicelo) |
| 19:11.25 | *** join/#maemo silviof (~silviof@unaffiliated/silviof) |
| 19:16.39 | *** join/#maemo cyteen (~cyteen@230.178.93.209.dyn.plus.net) |
| 19:56.24 | *** join/#maemo tmp_512 (5891430e@gateway/web/freenode/ip.88.145.67.14) |
| 20:06.00 | atk | I have a used n900, what should I do with it first? |
| 20:06.29 | CatButts | probably update to latest official firmware |
| 20:06.39 | CatButts | and install community something something |
| 20:06.48 | CatButts | https://wiki.maemo.org/Community_SSU |
| 20:08.29 | opt1mus | atk: and sand off the prongs on the charging cable you intend to use with it |
| 20:09.11 | opt1mus | the two little spring loaded prongs are sometimes quite proud and you risk breaking the charging port of the board |
| 20:09.14 | atk | I actually have already been using a N900 for a year I just want to get a clean N900 so I can move my current N900 stuff onto it because my current N900 has a dead keyboard |
| 20:09.18 | opt1mus | oh |
| 20:09.41 | atk | and I've already soldered down the charging port of my current one... that's something I need to do for the new one |
| 20:09.56 | opt1mus | nice |
| 20:09.57 | atk | and remove the shitty RTC battery |
| 20:10.16 | atk | whoops - I think I just accidentally murdered a small spider |
| 20:10.33 | opt1mus | I got some polar cell batteries, they're working quite well |
| 20:10.35 | atk | but yes, the thing I was specifically interested in is how do I flash these things back to "stock" |
| 20:10.50 | atk | yes, I have a polarcell in my current n900 and it was a great purchase |
| 20:11.09 | atk | I think I might get another one for when I forget to charge the phone for a few days and it dies. |
| 20:11.30 | opt1mus | I bought two and an external charger |
| 20:11.40 | atk | yes, the external charger is a must |
| 20:11.44 | Vajb | ~lazyflash |
| 20:11.45 | CatButts | hey guise, unrelated question |
| 20:11.59 | Vajb | meh |
| 20:12.19 | atk | Vajb: joerg has already pointed me towards that one, I'm going to read it in a second |
| 20:12.21 | CatButts | NiMh battery leak produces base Ph, yes? |
| 20:12.28 | atk | Vajb: but he told me to ask here anyway |
| 20:12.33 | CatButts | and not acid |
| 20:14.01 | atk | CatButts: "NiMH cells have an alkaline (basic) electrolyte..." from wikipedia |
| 20:14.07 | atk | so probably yes |
| 20:14.22 | CatButts | ah, good |
| 20:14.36 | CatButts | I am cleaning gunk from unrelated device with vinegar |
| 20:16.23 | KotCzarny | atk: you probably know about: |
| 20:16.25 | KotCzarny | ~flashing |
| 20:16.26 | infobot | rumour has it, maemo-flashing is http://wiki.maemo.org/Updating_the_tablet_firmware, or - on linux PC - download&extract http://maemo.cloud-7.de/maemo5/patches_n_tools/maemo-my-private-workdir.tgz, cd into it, do sudo ./flash-it-all.sh; or see ~flashing-cmdline, or see ~lazyflashing |
| 20:16.28 | KotCzarny | ~cssu |
| 20:16.28 | infobot | well, cssu is http://wiki.maemo.org/Community_SSU, or (Community Seamless Software Update) |
| 20:18.53 | DocScrutinizer05 | atk: do NPT sand off anything on USB plugs! |
| 20:19.44 | DocScrutinizer05 | aah, you coverd that already |
| 20:21.10 | atk | I'll just solder the thing down |
| 20:21.19 | CatButts | lol |
| 20:21.31 | atk | I have enough solder to cover the surface of the earth probably... |
| 20:22.00 | DocScrutinizer05 | ~usbfix |
| 20:22.00 | infobot | methinks usbfix is http://talk.maemo.org/showthread.php?t=75920 - and **NEVER** use epoxy (unless you want to seal your device for underwater), or https://www.youtube.com/watch?v=fYz5nIHH0iY#t=1866, you will basically need two irons: a small good one (or better hot-air reflow) and a 60+ Watt |
| 20:22.19 | CatButts | can solder and duct tape solve lovelife? |
| 20:23.26 | CatButts | So what's a quick list of things to do with a 'new' n900? I'll start |
| 20:23.45 | CatButts | 1. be a creep and look through previous owner's stuff |
| 20:24.02 | CatButts | 2. format and flash latest firmware |
| 20:24.05 | CatButts | 3. CSSU |
| 20:24.15 | CatButts | 4. sand them cables |
| 20:24.18 | atk | DocScrutinizer05: so basically, expose some copper on the ground plane next to the USB connector and then blob the fuck out of it |
| 20:24.20 | atk | got it :D |
| 20:24.21 | CatButts | 5. ???????? |
| 20:24.54 | CatButts | 5. Sturdy the fuck out of m-usb female |
| 20:25.00 | CatButts | 6. ???????? |
| 20:26.26 | *** join/#maemo troulouliou_div2 (~troulouli@unaffiliated/troulouliou-div2/x-0271439) |
| 20:26.28 | *** join/#maemo TheKit (~nekit@178.45.145.24) |
| 20:26.46 | atk | the libgles1 and libgles2 in the repos - that's from the SGX SDK? |
| 20:27.32 | Maxdamantus | Install u-boot, just in case you still lose the USB port |
| 20:28.29 | DocScrutinizer05 | I guess so, yes |
| 20:29.11 | atk | spider update: the spider is alive |
| 20:29.11 | DocScrutinizer05 | well, actually from Nokia, though I guess they also used that SDL to produce them |
| 20:29.22 | DocScrutinizer05 | \o/ :-D |
| 20:29.49 | CatButts | what's u-boot |
| 20:30.00 | Maxdamantus | ~u-boot |
| 20:30.00 | infobot | it has been said that u-boot is http://talk.maemo.org/showthread.php?t=81613 |
| 20:30.24 | DocScrutinizer05 | wonders if "keep a jam jar in freezer, to catch spiders in it" would qualify for a patent |
| 20:31.08 | atk | descriptions of wheels have qualified for patents before |
| 20:31.09 | atk | so why not |
| 20:31.11 | DocScrutinizer05 | ~factinfo u-boot |
| 20:31.11 | infobot | DocScrutinizer05: there's no such factoid as u-boot |
| 20:31.37 | DocScrutinizer05 | ~factinfo #maemo u-boot |
| 20:31.37 | infobot | #maemo u-boot -- created by DocScrutinizer51 <~lagrange@openmoko/engineers/joerg> at Mon Oct 29 12:36:32 2012 (1838 days); it has been requested 5 times, last by Maxdamantus, 1m 37s ago. |
| 20:31.46 | DocScrutinizer05 | aaah that's why |
| 20:33.25 | CatButts | https://i.imgur.com/GVI2n.jpg |
| 20:33.30 | CatButts | speaking of patents |
| 20:33.58 | DocScrutinizer05 | atk: the -26 centigrade nicely slow down spiders when you catch them by placing the jar over them on the wall. No escaping when getting the jar off the wall, after 20s |
| 20:34.16 | atk | I just pick them up with my hands |
| 20:34.18 | atk | it's easier |
| 20:34.26 | DocScrutinizer05 | *shudder* |
| 20:36.52 | Maxdamantus | usually just leaves them be. |
| 20:37.21 | Maxdamantus | will occasionally get rid of webs though. |
| 20:39.59 | atk | DocScrutinizer05: the md5sums don't match with any of the files provided by the SDK so I presume either they're from an older SDK version or they're custom nokia ones |
| 20:40.27 | *** join/#maemo Venemo (~venemo@fedora/Venemo) |
| 20:42.49 | DocScrutinizer05 | the SDK aiui provides source code interface stubs for including a binary blob into them and provide that with an ABI - so the files in the SDL would never show up unmodified on the target |
| 20:43.17 | DocScrutinizer05 | SDK even |
| 20:44.46 | atk | the SDK also provides the GLES libs |
| 20:45.09 | DocScrutinizer05 | and yeah, quite possibly Nokia had better docs for the PowerVR so they could even patch the blobs |
| 20:45.24 | atk | hmm, indeed |
| 20:45.34 | atk | well, when I get this compiling against mainline I'll try all manner of combinations |
| 20:46.00 | DocScrutinizer05 | also note that the latest SDK says "hardfloat" or somesuch, while maemo is softfp |
| 20:46.48 | atk | hmm, interesting |
| 20:47.04 | atk | does the n900 cpu support hardfloat? |
| 20:47.04 | DocScrutinizer05 | I'm not sure how much it would matter when the PVR libs use hardfp |
| 20:47.17 | DocScrutinizer05 | err yes |
| 20:47.19 | atk | it would probably make them a bit faster |
| 20:47.27 | DocScrutinizer05 | marginally |
| 20:48.24 | DocScrutinizer05 | even softfp in the end uses hardware "float coprocessor", just the parameter passing is via stack and a lib instead of coproc registers and direct function calls |
| 20:49.16 | CatButts | oh I have been reminded |
| 20:49.16 | DocScrutinizer05 | the lib then calls the float coproc as well on all architectures that support it |
| 20:49.29 | CatButts | n900 CPU and float |
| 20:49.35 | CatButts | vs integer |
| 20:49.44 | Maxdamantus | I don't think armel would accurately be described as softfp. |
| 20:49.54 | CatButts | on a nintendo DS, ou would go for integer |
| 20:50.18 | Maxdamantus | softfp is something you can enable when compiling something, where the compiler will use software implementation of floating point operations. |
| 20:50.49 | CatButts | that's probably slow |
| 20:51.03 | DocScrutinizer05 | indeed using reals/floats for GFX is ... *MEH!*, 32 or even 64bit integers should be absolutely sufficient and way faster |
| 20:51.04 | CatButts | compared to programming straight up int |
| 20:51.07 | Maxdamantus | Indeed, but it might be required for portability. |
| 20:51.47 | CatButts | so, n900's CPU cannot do float, yes? |
| 20:51.52 | Maxdamantus | If you're targeting a system that doesn't have any floating point operations, it might be better to just implement suitable fixed-point arithmetic. |
| 20:51.53 | DocScrutinizer05 | it can |
| 20:52.01 | Maxdamantus | Yes, N900 does both 32-bit and 64-bit floats. |
| 20:54.49 | DocScrutinizer05 | see http://www.ti.com/product/OMAP3530/description |
| 20:56.30 | DocScrutinizer05 | also see http://paste.opensuse.org/68509984 |
| 20:57.08 | DocScrutinizer05 | vfp and vfpv3 |
| 20:57.31 | DocScrutinizer05 | and neon on top |
| 20:57.37 | CatButts | >JazelleĀ® RCT Execution Environment Architecture |
| 20:58.16 | CatButts | oh hey, it's the NDA proprietary ARM java doohickey thing |
| 20:58.36 | CatButts | typically used for accelerating J2ME |
| 20:58.39 | DocScrutinizer05 | total crap, I heard |
| 20:58.46 | DocScrutinizer05 | slower than CPU to run java |
| 20:59.11 | CatButts | I am familiar with it from Sony Erricsson phones |
| 20:59.47 | Maxdamantus | I imagine the advantage would be potential power saving in not having to compile short-lived code. |
| 21:00.13 | CatButts | j2me apps run faster |
| 21:01.28 | Maxdamantus | Faster specifically when running code in the interpreted mode? |
| 21:02.46 | CatButts | faster when running intensive code that does stuff like render realtime audio |
| 21:03.03 | Maxdamantus | I'm not familiar with the ARM Java stuff, but I would imagine it would potentially be faster before the JIT actually compiles the code. |
| 21:03.05 | CatButts | there's no api for realtime audio, so it's a hack |
| 21:03.26 | Maxdamantus | Right, if it's something intensive, the JIT should compile it into native code anyway. |
| 21:03.51 | Maxdamantus | on a normal system, the threshold usually has to do with 10k iterations. |
| 21:04.10 | Maxdamantus | so eg, once a method has been called 10k times, it will probably be compiled to native code. |
| 21:04.43 | DocScrutinizer05 | well, I guess it depends on the SoC, no? |
| 21:04.52 | Maxdamantus | before that, it will just be interpreted .. so there'll be a while loop that looks something like: while(i = get_next_instruction()) run_instruction(i); |
| 21:05.32 | Maxdamantus | If you could replace that while loop with something equivalent in the CPU, that seems like a potential advantage. |
| 21:05.52 | DocScrutinizer05 | I've been told that OMAP3530 is faster on genuine ARM CPU than on Jazelle with running java |
| 21:11.11 | Maxdamantus | I imagine the better way to make things faster is to just put compiled code onto persistable storage. |
| 21:11.17 | Maxdamantus | ie, use actual binaries with native code in them. |
| 21:11.45 | Maxdamantus | which is what normally happens when your programs are written in C/C++/Rust/Go/whatever |
| 21:14.04 | Maxdamantus | iirc JDK9 has some support for AOT compilation too. |
| 21:14.55 | Maxdamantus | So you can compile a .so that you provide to the VM alongside your classes. |
| 21:19.58 | *** join/#maemo geaaru (~geaaru@host87-9-dynamic.183-80-r.retail.telecomitalia.it) |
| 21:34.58 | *** join/#maemo florian (~florian_k@Maemo/community/contributor/florian) |
| 21:55.49 | opt1mus | DocScrutinizer05: I don't see the issue with modifying a microUSB male in such a way, they're dirt cheap |
| 21:56.24 | opt1mus | if you leave them as is then you can run the risk of pulling at the USB port on the N900 too much |
| 21:56.49 | opt1mus | which would you rather? |
| 21:57.21 | opt1mus | a 2 bit cable, or bugger up a rare phone |
| 21:58.33 | DocScrutinizer05 | a) the plugs need a 100 mating cycles then they get softer b) sanding/filing causes nasty abrassive dust and surfaces that both can damage the receptacle when plugging that sanded plug c) pulling the plug never caused any hard, it's bending the plug while pulling it that does |
| 21:59.35 | DocScrutinizer05 | s/any hard/any harm/ |
| 22:00.17 | opt1mus | I think quite a few nonfunctioning USB ports on N900s disagree, to each their own |
| 22:01.09 | DocScrutinizer05 | and finally, razher than filing away the latches, you can simply push them inwards slightly with your fingernails, to make them a tad softer |
| 22:01.54 | opt1mus | hmm, perhaps bending them if that's possible would be favorable |
| 22:02.04 | opt1mus | if they can be made to remain less proud |
| 22:02.10 | DocScrutinizer05 | I used ~40 N900 and had no issues with ISB on a single one of them |
| 22:02.33 | DocScrutinizer05 | USB* |
| 22:03.18 | DocScrutinizer05 | I seen a dozen times how users abused their devices while unplugging USB |
| 22:04.06 | DocScrutinizer05 | as a general rule, never use both hands for unplugging, correct unplugging is done with two fingers of one hand basically |
| 22:04.20 | opt1mus | I baby mine, so it's never parted company from the board, so perhaps I'll never find out personally how much force is required to do so |
| 22:06.09 | opt1mus | maybe I've just been unlucky in having microUSB cables that've had overprominent prongs on them |
| 22:12.07 | DocScrutinizer05 | some are a bitch initially, yes. Pencil and a 10 minutes of prying out the plug with fingernail and then remating it fix this |
| 22:12.57 | DocScrutinizer05 | you can easily do 20 to 30 partial (1 - 2mm) unplug+replug cycles per minute |
| 22:13.59 | DocScrutinizer05 | use a soft pencil (HB, B...) to 'lubricate' the exterior of the plug a bit |
| 22:14.31 | opt1mus | interesting technique, it's not like I use a wood rasp on it though |
| 22:14.51 | opt1mus | ;} |
| 22:15.56 | DocScrutinizer05 | but the most important thing is to learn how to correctly unplug: grab the plug on *small* sides with your thumb and digit finger so close to the case that you can "roll" your fingertips against the case to pry out the plug |
| 22:16.42 | opt1mus | indeed |
| 22:18.37 | DocScrutinizer05 | alternatively insert fingernail between plug and case and pry out plug with your nail |
| 22:19.07 | DocScrutinizer05 | again bysupporting the finger(nail) against the case |
| 22:20.08 | DocScrutinizer05 | NEVER hold case with one hand and _pull_ the plug with the other hand, creating all the force out of your elbows |
| 22:20.31 | DocScrutinizer05 | or shoulders rather |
| 22:21.45 | DocScrutinizer05 | you must be avle to unplug using one hand only, while the device sints on a pillow or other soft surface and doesn't get pushed into it |
| 22:22.03 | DocScrutinizer05 | you must be able to unplug using one hand only, while the device sits on a pillow or other soft surface and doesn't get pushed into it |
| 22:25.39 | opt1mus | is this pillow made from silk, or Egyptian cotton? |
| 22:25.52 | opt1mus | I jest... |
| 22:44.27 | *** join/#maemo cyteen (~cyteen@230.178.93.209.dyn.plus.net) |
| 23:48.22 | *** join/#maemo Kabouik_ (~kabouik@176.67.168.147) |
