highaltitude.log.20151216

[00:00] <Ian_> Now wouldn't the export restrictions limit it's use on a large floating HAB, by definition.
[00:01] <Ian_> I believe that it's probably a current hog too.
[00:01] <SpeedEvil> People have used sat-stuff before
[00:01] <Ian_> I believe that the answer to your question is yes, but not many and I don't know who
[00:01] <Ian_> Too expensive for the North Sea . . .
[00:02] <mbales_> roger, it seems like it could be an interesting tool, if you can get it back and it works above 60k feet
[00:02] <mbales_> Ive used the SPOT trackers and they stop all updates at 60k
[00:03] <SpeedEvil> I vaguely recall Laurenceb considered DIY sat-trackers.
[00:03] <Ian_> I seem to recollect reading some blurb about the Rock Block, probably MK1, and it seemed 'fragile'.
[00:03] <chris_99> mbales_, is that just because it has a GPS though?
[00:04] <Ian_> Despite it's name.
[00:04] <mbales_> it very well may be
[00:04] <mbales_> i should test if it will transmit without a gps fix
[00:04] <mbales_> though im not sure how ill get an iridium fix without a gps fix
[00:05] DL1SGP (~felix64@dhcp29.signon1.dk.beevpn.com) left irc: Ping timeout: 256 seconds
[00:05] <Ian_> 60k(m) for Spot, isn't that optimistic. I think you may mean 12Km
[00:05] <mbales_> sorry, used America Units
[00:05] <mbales_> i mean Freedom Units
[00:05] <Ian_> How long is you Freedom Unit bit of string?
[00:06] <mbales_> about 4 hogs heads
[00:06] <Ian_> When I launch, I will paint a large black SPOT on the package and declare that I am using a spot for visual tracking.
[00:06] DL1SGP (~felix64@dhcp20.signon3.dk.beevpn.com) joined #highaltitude.
[00:07] <mbales_> black / orange would make it the correct color for a SPOT
[00:17] Ian_ (4d66af83@gateway/web/freenode/ip.77.102.175.131) left #highaltitude.
[00:17] Ian_ (4d66af83@gateway/web/freenode/ip.77.102.175.131) joined #highaltitude.
[00:20] <mbales_> downside is it costs $1 (USD) per KB
[00:21] gonzo_nb (~gonzo@host-92-6-244-28.as43234.net) left irc: Remote host closed the connection
[00:23] LazyLeopard (~irc-clien@88.97.98.120) left irc: Quit: Now QRT
[00:23] bertrik (~quassel@rockbox/developer/bertrik) left irc: Remote host closed the connection
[00:30] andyvk5 (~andyvk5@zedm.net) joined #highaltitude.
[00:31] andyvk5_ (~andyvk5@zedm.net) joined #highaltitude.
[00:33] andyvk5 (~andyvk5@zedm.net) left irc: Read error: Connection reset by peer
[00:40] michal_f (~michal_f@91.146.241.130) left irc: Quit: HydraIRC -> http://www.hydrairc.com <- In tests, 0x09 out of 0x0A l33t h4x0rz prefer it :)
[00:59] <Laurenceb> lulwut http://forum.nasaspaceflight.com/index.php?topic=39004.msg1457766#msg1457766
[01:00] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Quit: Leaving
[01:24] <mbales_> ha
[01:49] DL7AD (~sven@p4FD4204E.dip0.t-ipconnect.de) left irc: Ping timeout: 246 seconds
[01:50] DL7AD (~sven@p4FD42099.dip0.t-ipconnect.de) joined #highaltitude.
[01:51] <DoYouKnow> hi. I finally got around to testing my theory about the MON-SIG... apparently this MON-SIG command reports back a 256-point fft on the ublox7
[01:51] <DoYouKnow> (usb dongle)
[01:51] <DoYouKnow> around the gps band
[01:52] <DoYouKnow> well, probably the gps band
[01:52] <DoYouKnow> so, I want to plot this, so I need to extract the new message type I enabled 0x0a 0x1c
[01:53] <DoYouKnow> here is an example fft that I manually plotted: http://hosting.datacopia.com/publish/20151222.233246.3bff33e0-b244-487d-a3a3-ebcb5196bc6c.png
[01:53] <DoYouKnow> so I want to extract this new message type with rtklib
[01:54] <DoYouKnow> any idea how to do this with rtklib?
[01:55] <DoYouKnow> this is because I want to generate a scrolling 256 x t waterfall
[01:57] <DoYouKnow> Hello?
[01:57] <adamgreig> it's like 2am uk time so you might have to wait a bit
[01:58] <DoYouKnow> ok
[01:58] <adamgreig> not sure many people have played with rtklib here either
[01:58] <adamgreig> sounds like what you want should be ok in general though
[01:58] <DoYouKnow> ok thnx
[02:00] K9JKM (~chatzilla@c-73-246-133-246.hsd1.il.comcast.net) joined #highaltitude.
[02:01] <Laurenceb> https://www.youtube.com/watch?feature=player_embedded&v=wLHCuzW3-uA
[02:01] <Laurenceb> erm not that
[02:01] <Laurenceb> this
[02:01] <Laurenceb> http://www.spacex.com/careers/position/8164
[02:01] Action: Laurenceb reorders his troll links
[02:01] SpeedEvil (~quassel@tor/regular/SpeedEvil) left irc: Quit: No Ping reply in 180 seconds.
[02:02] SpeedEvil (~quassel@tor/regular/SpeedEvil) joined #highaltitude.
[02:02] <Laurenceb> trolls all the way down
[02:23] Action: Ian_ gets Marigolds for a busy day in the SpaceX kitchens
[02:25] <Ian_> Disposables would save on pensions and expensive security vetting for a dishwasher
[02:36] DL1SGP (~felix64@dhcp20.signon3.dk.beevpn.com) left irc: Ping timeout: 260 seconds
[02:37] DL1SGP (~felix64@dhcp49.signon1.dk.beevpn.com) joined #highaltitude.
[02:54] K9JKM (~chatzilla@c-73-246-133-246.hsd1.il.comcast.net) left irc: Quit: ChatZilla 0.9.92 [Firefox 42.0/20151029151421]
[03:56] Laurenceb (~Laurence@host86-179-250-250.range86-179.btcentralplus.com) left irc: Ping timeout: 272 seconds
[04:25] Flutterb1t (~flutterba@unaffiliated/day) joined #highaltitude.
[04:29] Flutterbat (~flutterba@unaffiliated/day) left irc: Ping timeout: 256 seconds
[04:29] Nick change: Flutterb1t -> Flutterbat
[06:21] es5nhc (~tarmo@108-40-71-217.sta.estpak.ee) joined #highaltitude.
[06:50] ljenkins_ (~ljenkins@2001:1948:212:1030:b1c7:ab2c:6b85:d4de) left irc: Read error: Connection reset by peer
[06:51] ljenkins_ (~ljenkins@2001:1948:212:1030:912b:27ea:de5:abf4) joined #highaltitude.
[07:27] fab4space (~fab4space@AMontpellier-656-1-416-68.w90-0.abo.wanadoo.fr) left irc: Ping timeout: 240 seconds
[08:26] ipdove (~ipdove@interclub.plus.com) joined #highaltitude.
[08:33] fab4space (~fab4space@109.237.242.98) joined #highaltitude.
[08:46] <UpuWork> BARC are flying again today
[08:54] <mfa298> Apparently bacon is good, science says so:
[08:54] <mfa298> http://www.telegraph.co.uk/news/earth/environment/12052711/Lettuce-worse-than-bacon-for-the-environment-scientists-claim.html
[08:57] <AndyEsser> o/
[08:57] DL1SGP (~felix64@dhcp49.signon1.dk.beevpn.com) left irc: Ping timeout: 272 seconds
[08:59] DL1SGP (~felix64@dhcp6.signon3.dk.beevpn.com) joined #highaltitude.
[09:18] <Chimpusmaximus> UpuWork: any idea of time?
[09:26] <AndyEsser> Chimpusmaximus: it's currently 9:26:36
[09:26] <AndyEsser> :)
[09:27] <UpuWork> "11 ish"
[09:27] <Chimpusmaximus> ;-)
[09:27] <Chimpusmaximus> Cheers
[09:29] Hiena (~boreger@81.93.195.181.datatrans.hu) joined #highaltitude.
[09:30] <Chimpusmaximus> Upework: does the habsupplies shutdown over xmas and new year?
[09:30] <UpuWork> lol as if
[09:31] <Chimpusmaximus> good good
[09:31] <AndyEsser> :)
[09:31] <AndyEsser> Good, I may be placing orders :)
[09:41] michal_f (~michal_f@84-10-62-166.static.chello.pl) joined #highaltitude.
[09:43] T-MaN (~thomas@217.21.233.78) joined #highaltitude.
[09:51] <fsphil> quick arm question. gcc has produced this: str r0, [sp, #4]
[09:51] <fsphil> sp is pointing to the end of ram at this point, isn't it trying there to store r0 to an invalid location outside of ram?
[09:53] <zyp> why would sp be pointing to the end of ram?
[09:54] <fsphil> it's a descending stack
[09:54] <zyp> sure, but why would it try to store something on the stack before allocating space?
[09:54] <zyp> pastebin the whole function
[10:03] Lunar_Lander (~kevin@p54889688.dip0.t-ipconnect.de) joined #highaltitude.
[10:12] <fsphil> http://pastebin.com/raw/pbczN79J
[10:12] <fsphil> excuse the hacky nature of this code, I'm just messing around
[10:13] <fsphil> it writes the '!' to the uart, then nothing
[10:13] <Hiena> Never apologize for your codes quality. That makes you look more unprofessional than a whacked-hacked, hello.c
[10:14] <adamgreig> fsphil: have you stepped through in a debugger, checked where sp is pointing exactly?
[10:15] <fsphil> not yet. I've not used gdb on a micro yet. good time to start eh
[10:15] <adamgreig> yep :p
[10:15] <fsphil> I'm presuming it's loaded SP with the first value in my vector table
[10:16] <zyp> fsphil, I mean't the assembly for the function
[10:16] <zyp> -'
[10:16] <fsphil> ah, one sec
[10:16] <zyp> but yeah, source helps too
[10:17] <fsphil> https://www.sanslogic.co.uk/disassembly.txt
[10:18] <zyp> remove __attribute__ ((naked))
[10:18] <zyp> as per https://gcc.gnu.org/onlinedocs/gcc/ARM-Function-Attributes.html#ARM-Function-Attributes "Only basic asm statements can safely be included in naked functions"
[10:19] <zyp> so that's probably why it's not making proper allocations on the stack
[10:20] <fsphil> hmm. bad idea writing the startup code in C then?
[10:20] <zyp> no, that's fine, a non-naked function should work perfectly fine as a reset vector
[10:21] <zyp> the stack gets initialized by hardware, so a function prologue starting to use the stack at the function entry is perfectly fine
[10:24] <fsphil> so it starts now with push {r0, r1, r2, r4, r5, r6, r7, lr}
[10:24] <fsphil> that'll stop that str instruction writing outside ram anyway
[10:25] <fsphil> though seems a shame to waste those bytes :)
[10:26] <fsphil> still overwriting the last entry in the stack, which is odd
[10:26] <fsphil> if [sp, #4] does what I think it does anyway
[10:26] <zyp> it does
[10:26] <adamgreig> step it in the debugger
[10:27] <zyp> how does the disassembly look now?
[10:29] <fsphil> same, but with the addition of that push {} at the start: https://www.sanslogic.co.uk/disassembly2.txt
[10:30] <adamgreig> huh, no pop at the end?
[10:30] <fsphil> yeah noticed that. guess it noticed the function never returns
[10:31] <adamgreig> if you compile with gcc -s -O0 bla.c what do you get?
[10:31] <adamgreig> should get annotated assembler at least
[10:31] <UpuWork> .flights
[10:31] <SpacenearUS> 03UpuWork: Current flights: 03BARC 10(25f4)
[10:31] <zyp> huh, still no stack allocation
[10:32] <zyp> that's really weird
[10:32] <zyp> which optimization level is that?
[10:33] <fsphil> Os
[10:33] <fsphil> just trying O0 now
[10:35] <fsphil> ah, big -S adamgreig
[10:35] <adamgreig> oh yep oops
[10:36] <fsphil> https://www.sanslogic.co.uk/assembly.txt
[10:37] <fsphil> that is weird
[10:37] <fsphil> oh
[10:37] <fsphil> wrong gcc, duh
[10:37] <adamgreig> lol
[10:37] <adamgreig> arm-none-eabi-gcc :P
[10:39] <fsphil> refresh. *ahem*
[10:40] <fsphil> quite a bit longer
[10:42] <adamgreig> quite different too lol
[10:42] <adamgreig> no more str to sp?
[10:43] englishman (~englishma@alcohol.dtfuhf.com) left irc: Ping timeout: 256 seconds
[10:43] <fsphil> doesn't seem to have. only access to sp is the sub and add at the start of reset_handler()
[10:45] <fsphil> that's probably running. I'm not near the board to check the LEDs :)
[10:45] <fsphil> oh, I think I left the serial port connected
[10:46] <fsphil> yeah, printing all the 'a's. that's progress
[10:46] <adamgreig> so it.. works with -O0 and breaks with -Os?
[10:47] <adamgreig> because if so it's time for game #2: find where you're using undefined behaviour
[10:47] <fsphil> working with -Os now too. only difference is its not naked
[10:47] <adamgreig> ah ok
[10:47] <adamgreig> cool
[10:47] <adamgreig> normally i put my code in main() instead of writing my own reset handler :P
[10:49] <fsphil> hah. thought I'd see how hard it was...
[10:49] <fsphil> so, with ((naked)) it works fine with -O0 and breaks with -Os
[10:50] <adamgreig> yea but OB
[10:50] <adamgreig> UB
[10:50] <adamgreig> lol
[10:51] <fsphil> guess the naked attribute just strips the setup code, doesn't change how it compiles the actual code. it just assumes everything is happy
[10:56] <fsphil> https://github.com/libopencm3/libopencm3/blob/master/lib/cm3/vector.c#L63
[10:57] <fsphil> they seem to be doing a similar thing
[10:58] <AndyEsser> fsphil: one thing I noticed, but is very very unlikely to cause your problems.... please change "int i;" to "int i = 0;" :P
[10:59] <fsphil> AndyEsser: for(i = 0; ... :)
[11:00] <AndyEsser> That's why I said it was unlikely, as the first thing you do is assign it - but it upsets my karma ;)
[11:00] T-MaN (~thomas@217.21.233.78) left irc: Remote host closed the connection
[11:02] drsnik (~drsnik@gate3.ima.cz) joined #highaltitude.
[11:02] drsnik (~drsnik@gate3.ima.cz) left irc: Remote host closed the connection
[11:03] <fsphil> you do it for all variables?
[11:03] englishman (~englishma@alcohol.dtfuhf.com) joined #highaltitude.
[11:03] <fsphil> gcc gets shouty if I do it wrong
[11:03] <AndyEsser> Yes - always initialise variables on declaration
[11:04] <AndyEsser> in C not initialising variables is technically undefined behaviour - in reality the compiler is free to do whatever it likes for it. I'm just wondering if the for loop might do something weird withit before assigning the value
[11:04] <AndyEsser> but again _highly_ unlikely to be causing you any problems
[11:04] drsnik (~drsnik@gate3.ima.cz) joined #highaltitude.
[11:04] <fsphil> only undefined if you read it first
[11:04] g0hww (~g0hww@46.18.105.34) joined #highaltitude.
[11:05] <fsphil> if the compiler tired to do something with i before it gets to i = 0, it's a bug in the compiler
[11:06] <AndyEsser> well... gcc isn't the best compiler around :P
[11:06] <fsphil> I did hit a very similar bug in avr-gcc when writing ssdv
[11:07] <fsphil> one variable in one function never got initialised
[11:07] <fsphil> even though it was in the C code
[11:07] <AndyEsser> just reading up now
[11:07] <eroomde> gcc is good! at least compared to other ones i've tried
[11:07] <AndyEsser> eroomde: gcc is good (at least in relation to the MS C/C++ compiler)
[11:07] <fsphil> someone here confirmed it was a gcc bug
[11:07] <AndyEsser> but clang is just blowing them all away in recent years
[11:08] <fsphil> it's been fixed since
[11:08] <eroomde> i've not tried things that have a good reputation for x86 optimisation though, like intel's
[11:08] <nick_> Some languages have the good feature that all types have a default value, which is set if you don't do it yourself (often zero/null)
[11:08] englishman (~englishma@alcohol.dtfuhf.com) left irc: Ping timeout: 246 seconds
[11:08] <AndyEsser> nick_: C standard doesn't
[11:08] <AndyEsser> only for type char
[11:08] <nick_> Yes, "some languages" isn't including C/C++.
[11:08] <AndyEsser> :)
[11:09] <nick_> Go is the main one that I actually use this feature in.
[11:09] <eroomde> i've used the clang static analyser before, and i think it's the default compiler for os stuff on osx now
[11:09] <AndyEsser> fsphil: anyway, I just got into the habit of initialising when declaring since I'm sure it's tripped me up once or twice before
[11:10] <eroomde> there was a thing of nastry tricks to play on c programmers doing the rounds a bit alo
[11:10] <eroomde> the one i really liked was:
[11:11] <eroomde> #define volatile
[11:11] <AndyEsser> change your ; to the greek letter?
[11:11] <AndyEsser> ha
[11:11] <AndyEsser> o god... I remember seeing a few years ago, in the Google Android OS source code, something like
[11:11] <dbrooke> I've ported code to a C environment which didn't initialise BSS which was interesting before I realised
[11:11] <AndyEsser> #define FASLE false;
[11:11] <fsphil> static and global variables all get set to 0 in C
[11:11] <AndyEsser> #define FSALE fale;
[11:11] <fsphil> unless what dbrooke said happens :)
[11:12] englishman (~englishma@alcohol.dtfuhf.com) joined #highaltitude.
[11:12] <dbrooke> lots of arrays of pointers no longer NULL
[11:13] <AndyEsser> fsphil: yes, static are because you have to in order to compile. global would surprise me
[11:13] <dbrooke> required some judicious use of memset()
[11:13] <AndyEsser> and your int i is neither ;)
[11:13] <AndyEsser> dbrooke: ah, the C programmers greatest tool
[11:13] <AndyEsser> although
[11:13] <AndyEsser> memset caused insta-resets on my uC last night
[11:13] <AndyEsser> so had to loop over manually myself and set the bytes
[11:14] <AndyEsser> I suspect I may have calculated the size wrong though
[11:17] <dbrooke> as these were arrays then sizeof did the job
[11:18] <AndyEsser> well I had a char myArray[256][16];
[11:18] <AndyEsser> so would've thought len = sizeof(char) * 256 * 16;
[11:18] <AndyEsser> but apparently, no :P
[11:19] <fsphil> a 4k array in your 2k microcontroller? :)
[11:19] <AndyEsser> ha
[11:19] <AndyEsser> those numbers might not be accurate
[11:19] <AndyEsser> :P
[11:19] <fsphil> lol
[11:19] <AndyEsser> but yes... apparently so
[11:19] <Hiena> Magic!
[11:20] <AndyEsser> indeed
[11:20] <AndyEsser> however, that might've explained some of my other issues last night
[11:20] <AndyEsser> :P
[11:20] <SpacenearUS> New position from 03BARC after 03a day silence - 12http://tracker.habhub.org/#!qm=All&q=BARC
[11:20] <AndyEsser> that's an interesting flightpath :)
[11:22] <adamgreig> it's not UB to access an uninitialised local variable unless the type of the variable may contain a trap btw, which ints don't on arm
[11:23] <adamgreig> it's certainly not UB to declare a variable and later assign to it before use
[11:23] <AndyEsser> IT'S A TRAP!!
[11:23] <AndyEsser> adamgreig: I'm aware of the second bit, I was just trying to rule out any funky code doing something in the meantime to i
[11:29] <eroomde> i've just met someone whose job title is 'head of digital'
[11:30] <eroomde> digital what, they didn't specify
[11:30] <adamgreig> did you write "head of analogue" on your business card before handing it over?
[11:30] <dbrooke> let the compiler do the work - sizeof myArray
[11:31] <Chimpusmaximus> !dial BARC
[11:31] <SpacenearUS> 03Chimpusmaximus: Latest dials for 03BARC 10(25f4): none
[11:34] <AndyEsser> dbrooke: I've always been taught that only types should be passed to sizeof() not variables
[11:34] <AndyEsser> many many moons ago
[11:34] <AndyEsser> and I can't remember the reasoning
[11:34] <zyp> that doesn't make sense
[11:35] <dbrooke> Chimpusmaximus: 434.450 but I've not had a green yet
[11:35] <Chimpusmaximus> cheers
[11:35] <eroomde> i should do that thougb adamgreig
[11:35] <Chimpusmaximus> though i might have seen on CheshireSDR
[11:36] <Chimpusmaximus> thought
[11:36] Action: AndyEsser walks down the road to CheshireSDR and gives it a kick
[11:37] <Chimpusmaximus> So i must be blind
[11:38] <gonzo_> AndyEsser, isn't that what your are actually doing
[11:39] <mfa298> I think I came across something that suggested you need to be careful with sizeof with arrays as you will get differnt results from inside a function call (you'll get the size of a pointer rather than the size of the allocated memory)
[11:39] <mfa298> which might be where advice to use sizeof with types comes from.
[11:39] <gonzo_> I always assume that arrays are just pointers
[11:40] <gonzo_> so I'd pass the type, then multiplu by the defined size
[11:41] <AndyEsser> gonzo_: which was what I've always done as well
[11:41] <gonzo_> but for basic types (though I always typecast everythiung anyeay) you can use sizeof
[11:41] <SpacenearUS> New position from 03SP6NVB-11 after 0312 hours silence - 12http://tracker.habhub.org/#!qm=All&q=SP6NVB-11
[11:42] <gonzo_> not a clue what would happen these days with objects etc.
[11:42] <dbrooke> AndyEsser: there are potentially scope issues to be aware of, like trying to find size of an array in a function, when it's actually been passed as a pointer
[11:43] <AndyEsser> dbrooke: yea
[11:43] <dbrooke> oops, mfa298 just said that before I read back
[11:43] <gonzo_> I would either pass in the sizes as a param to the fn, or use common #def'ed values
[11:44] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[11:46] <SpacenearUS> New vehicle on the map: 03F6AGV - 12http://tracker.habhub.org/#!qm=All&q=F6AGV
[11:47] <michal_f> !flights
[11:47] <SpacenearUS> 03michal_f: Current flights: 03BARC 10(25f4)
[11:53] Hix_ (~hix@97e08719.skybroadband.com) left irc: Read error: No route to host
[11:53] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[11:55] <adamgreig> you can also pass sized arrays as parameters to functions
[11:55] <adamgreig> though gcc doesn't like it much :P
[11:56] Hix_ (~hix@97e08719.skybroadband.com) left irc: Read error: No route to host
[11:56] <adamgreig> arrays are not pointers, it's worth mentioning
[12:07] SushiKenBrown_ (~quassel@cmr-208-124-174-194.cr.net.cable.rogers.com) joined #highaltitude.
[12:07] SushiKenBrown (~quassel@208.124.174.194) left irc: Ping timeout: 272 seconds
[12:17] Davespice_ (~quassel@cpc88325-haye26-2-0-cust1840.17-4.cable.virginm.net) left irc: Quit: http://quassel-irc.org - Chat comfortably. Anywhere.
[12:17] Davespice (~quassel@cpc88325-haye26-2-0-cust1840.17-4.cable.virginm.net) joined #highaltitude.
[12:23] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) joined #highaltitude.
[12:26] <AndyEsser> !flights
[12:26] <SpacenearUS> 03AndyEsser: Current flights: 03BARC 10(25f4)
[12:27] <gonzo_> I have always treated array names as pointers
[12:27] <AndyEsser> same
[12:28] <gonzo_> so caName is a pointer direcx tly intercgangeable with &caName[0]
[12:28] <eroomde> char *this = "doesn't care how long this is"; // but; char this[6] = "well upset gcc i think" // though probably only with the right warnings
[12:28] <mfa298> whilst they might be technically different, they're often passed around as pointers and then treated like arrays again inside functions, which is when sizeof then starts to fail
[12:29] Geoff-G8DHE (~Geoff-G8D@geoffg8dhe.plus.com) joined #highaltitude.
[12:29] <gonzo_> i've always assumed that passing an array/struct on the stack, is just a pointer
[12:29] junderwood (~John@host86-166-189-20.range86-166.btcentralplus.com) joined #highaltitude.
[12:29] <junderwood> !dial barc
[12:29] <SpacenearUS> 03junderwood: Latest dials for 03BARC 10(25f4): none
[12:29] <fsphil> never tried this. would gcc warn if you passed an int a[10] to a function declared with a[5]
[12:29] <AndyEsser> eroomde: that will upset gcc because it's too long
[12:29] <adamgreig> eroomde: you can also `char this[] = "is fine and gcc will allocate enough space";`
[12:30] <gonzo_> have seen:
[12:30] <gonzo_> char *this[] = "doesn't care how long this is";
[12:30] <adamgreig> gonzo_: passing anything on the stack, the only difference is what type your code treats it as
[12:30] <AndyEsser> gonzo_: that seems weird
[12:30] <AndyEsser> since technically char* this[] is an array of char*
[12:30] <adamgreig> christ, don't do char *this[] = "bla"
[12:30] <AndyEsser> adamgreig: b l a could be valid pointers ;)
[12:30] <adamgreig> yea that won't do anything good
[12:30] <gonzo_> sorry, lose the *
[12:31] <gonzo_> an arrasy of pointers!
[12:31] <adamgreig> AndyEsser: even if they were, "bla" wouldn't be a valid initialisation for them
[12:31] <AndyEsser> char this[] = "blah blah"; is valid
[12:31] <AndyEsser> and I would then treat this as a pointer
[12:31] <AndyEsser> that was a bad variable name...
[12:32] <adamgreig> but `this` would be an array not a pointer
[12:32] <adamgreig> they are similar and different
[12:32] <gonzo_> the pc types/C++ people here seem top recon that passing arrays/pointers actually passes the whole thing on the stack. Not sure if that is true?
[12:32] <fsphil> we where talking about this recently. doesn't "char this[] = ...." have no guarantee of being null terminated?
[12:32] Geoff-G8DHE_ (~Geoff-G8D@geoffg8dhe.plus.com) left irc: Ping timeout: 260 seconds
[12:32] <adamgreig> if you later cast it to a pointer, (char*)this, or passed it to a function that took a pointer, then it becomes a pointer
[12:32] <adamgreig> fsphil: it will be null terminated if you used []
[12:32] <AndyEsser> fsphil: depends on compiler, usually static strings are auto-terminated in compilers I've seen
[12:32] <adamgreig> if you have [12] and pass it 12 characters in the string you won't get null termination (don't do that)
[12:32] <fsphil> ah that was it
[12:33] <adamgreig> `char s[5] = "hello"` is not null terminated
[12:33] <adamgreig> same as char s[] = {'h', 'e', 'l', 'l', 'o'}
[12:34] <fsphil> strings in C are a bit of a pain
[12:34] <adamgreig> lol
[12:34] <gonzo_> if preinit on an array, if I've declared the size, I have always used {'1','2','3'}, rather than the "123"
[12:34] <lz1dev> a bit...
[12:34] <adamgreig> char arrays in C, on the other hand :P
[12:35] <gonzo_> I don't think strings exist in C, only char arrays
[12:35] <adamgreig> string literals exist in C
[12:36] <gonzo_> only recently have ai had to even thing about them as strings. As I've always had to knife and fork my way throuygh with for loop etc. Only recently used the printf
[12:36] <adamgreig> and string.h is defined in the C standard too, so I think it's fair to say C has strings
[12:36] <gonzo_> have never has a system with enough spare prog space to incluse bloaty libs with string stuff
[12:37] <gonzo_> when I think about the standard lanugahe, I assume the bare compiler. Not any extra libs that happen to eb bundled in
[12:37] <gonzo_> that has lead to lots of arguments here
[12:38] <adamgreig> lol
[12:38] <gonzo_> and that exact argument above has happened
[12:38] <AndyEsser> for something like embedded programming
[12:39] <AndyEsser> as few #include <>'s as possible :)
[12:39] <adamgreig> ehhh
[12:39] <fsphil> stdint.h
[12:39] <AndyEsser> (within reason)
[12:39] <adamgreig> as soon as you find yourself writing your own memcpy you should really be using string.h
[12:39] <fsphil> everything else is optional :)
[12:39] <AndyEsser> adamgreig: yes apologies - hence the addendum of (within reason)
[12:40] <AndyEsser> but memcpy/memset etc are in stdlib.h not string.h, no?
[12:40] <adamgreig> nope
[12:40] <gonzo_> which lasted a whople day. Till I had to set up a test and prove that strings were not part of the language. Was then told I had to add the include for the lib, which took a few hours to find and it blew the flash limits.
[12:40] <adamgreig> memcpy and memset are both string.h
[12:40] <AndyEsser> interesting
[12:40] <gonzo_> all the mem fn's, had top write my own
[12:40] <adamgreig> gonzo_: not the C standard's fault you have a bloated implementation of the library, doesn't change that it's in the standard document ;)
[12:41] <AndyEsser> maybe I'm thinking of strlen/strcmp... which to me should still be in string.h
[12:41] <adamgreig> but that's the point - once you've written your own you might as well use someone else's well written one
[12:41] <adamgreig> get a libc designed for embedded systems
[12:41] <adamgreig> your linker removes functions you don't call
[12:41] <adamgreig> strlen is also string.h
[12:41] <adamgreig> as is strcmp..?
[12:41] <AndyEsser> "get a libc designed for embedded systems" <--- This. Do you know how many arguments I've had with people in the game's industry about including vanilla STL
[12:42] <gonzo_> was limited by the tool chain that was in place when i joined the project
[12:42] <adamgreig> sure, sometimes you have to just do your own thing for whatever reason
[12:42] <adamgreig> not C's fault though :P
[12:42] <gonzo_> (and tools that the vendor had gone bust/been brought out/lost the sources a decade before)
[12:42] <adamgreig> yea
[12:42] <AndyEsser> bbin15 grabbing lunch
[12:42] <adamgreig> embedded stuff rapidly descends into a nightmare :P
[12:42] <adamgreig> just stick with assembler >_>
[12:43] <fsphil> the leo method
[12:43] <gonzo_> it was more that I was being managed by PC types who have lived in infinite resource land/os land
[12:44] <adamgreig> mm
[12:44] junderwood (~John@host86-166-189-20.range86-166.btcentralplus.com) left irc: Quit: Leaving
[12:44] <lz1dev> infinite resource land :>>>
[12:45] <gonzo_> (of course that was because there were PC types who called themselves embedded, because the kit as in a box, not a desk top pc. But still a PC with windows....)
[12:46] <fsphil> seen that too often :(
[12:46] <lz1dev> http://www.dell.com/us/business/p/poweredge-r920/pd
[12:46] <lz1dev> :>
[12:46] DL7AD (~sven@p4FD42099.dip0.t-ipconnect.de) left irc: Quit: Leaving.
[12:47] <gonzo_> it is fun when you then spec to them some IO actions with acuracies req in usec
[12:47] <fsphil> new habhub server?
[12:47] <lz1dev> you wish
[12:47] <lz1dev> 96 dimms slots
[12:47] <dbrooke> part of a system I used to work on used XP embedded, luckily I worked on a different part but still had an RTOS
[12:48] DL7AD (~sven@p4FD42099.dip0.t-ipconnect.de) joined #highaltitude.
[12:48] ibanezmatt13 (~ibanezmat@host81-131-186-143.range81-131.btcentralplus.com) joined #highaltitude.
[12:48] <fsphil> we've a few machines with Win2000 embedded. I've yet to see the difference from normal 2k
[12:48] <Vaizki> only intel cpu I've been excited about for a while is the Xeon-D
[12:49] <gonzo_> I'm workjing on pc stuff now. But when they start talking to me I just pull the mong face, and tell them to keep talking and I'l;l stop them when I hear some words I understand
[12:49] <Vaizki> fsphil, there is zero difference except for the install afaik
[12:49] <UpuWork> it came with a read only file system fsphil
[12:50] <lz1dev> 'embedded'
[12:50] <fsphil> that's it?
[12:50] <Vaizki> well yes and that, you could run it from read only fs + ramdisk + compact flash or such
[12:50] <dbrooke> with XP embedded I think you could remove certain components but it still seemed to have lots of stuff I thought to be unnecessary
[12:50] <fsphil> it's installed on a plain old ide disk
[12:51] <lz1dev> oh yeah, windows has always had problem relating to storage
[12:51] <Vaizki> yea I have a R&S cellular analyzer with win2k embedded on it
[12:51] <lz1dev> for some reason it doesn't like anything thats not hdd
[12:52] <lz1dev> even today you can't do backups on a flash drive
[12:52] <lz1dev> which makes no sense
[12:53] <SpacenearUS> New position from 03WB8ELK-5 after 0315 hours silence - 12http://tracker.habhub.org/#!qm=All&q=WB8ELK-5
[12:55] <Vaizki> I believe you can run XP embedded from compact flash with FAT32 but NTFS is futile
[12:56] <lz1dev> basically dont use windows for embedded stuff
[12:57] <Vaizki> yea.. I wouldn't.. but then again I have a few devices that use it and they are fine
[12:57] <Vaizki> most surprising to me was that my FLIR thermal camera has windows inside
[12:58] <lz1dev> my condolences
[13:00] <Vaizki> naah it's great.. easy to hack because windows :D
[13:02] <fsphil> wait, that's a feature now?
[13:02] <fsphil> I've been so wrong all these years :)
[13:03] <Vaizki> it is for me as a user ;)
[13:03] <lz1dev> inb4 internet of windows things
[13:05] <Vaizki> resolution upgrade 80x60 -> 320x240, image noise reduced, additional measurement modes etc with a 5min hack..
[13:05] <Vaizki> yes, it's the same HW as a £5000 model but software crippled
[13:05] <fsphil> hah
[13:05] <fsphil> nice
[13:07] <Vaizki> someone should fly a raspi + lepton sensor :)
[13:09] <realborg> how to build a lepton sensor?
[13:09] <fsphil> with great difficulty. or buy one :)
[13:09] <fsphil> https://www.sparkfun.com/products/13233
[13:10] <chris_99> or the seek thermal has a higher res iirc
[13:11] <Vaizki> my camera has visual camera with thermal overlay and edge detection which makes it a joy to use compared to thermal only
[13:11] <Vaizki> example from flir web: http://www.flir.com/uploadedImages/Thermography_USA/Products/Ex-Series/Elect%20Panel%20msx.jpg
[13:12] <Vaizki> compared to thermal only http://www.flir.com/uploadedImages/Thermography_USA/Products/Ex-Series/Elect%20Panel%20thermal%20only.jpg
[13:13] <Vaizki> hmm that sparkfun board also says 80x60 resolution
[13:13] <Vaizki> I wonder if they actually made a new HW module with that resolution or if that is really 320x240 also
[13:13] <AndyEsser> fsphil: speaking of habhub server, where is it? and who maintains it?
[13:14] <adamgreig> http://habhub.org/
[13:14] <adamgreig> scroll down
[13:14] <fsphil> what adamgreig said
[13:14] <AndyEsser> o derp
[13:14] <AndyEsser> sorry
[13:14] <adamgreig> it's in manchester
[13:14] <adamgreig> maintained by various overworked sysadmins in #habhub :P
[13:14] <fsphil> haha
[13:15] <adamgreig> (not really, it's very low maintenance)
[13:15] <Vaizki> the links don't seem to work? :)
[13:15] <AndyEsser> you can't call yourself a sysadmin unless you're overworked ;)
[13:15] <adamgreig> lol indeed
[13:15] <adamgreig> we haven't actually put web pages on ceto or phorcys yet as haven't found any good pictures
[13:15] <adamgreig> http://kraken.habhub.org/ http://tiamat.habhub.org/
[13:16] <adamgreig> kraken and tiamat still exist though, just we virtualised them and put them onto ceto and phorcys
[13:16] <lz1dev> i can send you a picture of the rain currently coming down on the hut
[13:16] <fsphil> huh, no picture on tiamat
[13:16] <adamgreig> and by 'we' I mean DanielRichman did it all
[13:16] <mfa298> AndyEsser: So true. I'm just using my first day of Annual leave today for this year, The AL year only started in April.
[13:16] <Vaizki> so they run in a shed in lz1dev's yard? :)
[13:16] <lz1dev> its fairly close by
[13:16] <AndyEsser> mfa298: yea, I did similar things
[13:17] <AndyEsser> I used to work in education, and all the teachers were like "You going away over summer?"
[13:17] <adamgreig> lol, I think it's a fairly legit datac..ontainer
[13:17] <fsphil> haha
[13:17] <AndyEsser> "Erm no.. I'll be here. it's the only time I can do any proper work because you're incapable of putting paper in a printer yourself"
[13:17] <fsphil> on a roof somewhere
[13:17] <AndyEsser> hehe
[13:18] <fsphil> so habhub is fairly flood proof, of the water kind anyway
[13:18] <Vaizki> I bought a really cheap low voltage xeon server with 24GB memory from a UK refurb dealer and slapped some 2TB sata disks into it, put it in the company lab.. :)
[13:19] <fsphil> I was thinking of claiming some of our old servers for home, but the noise would drive me mad I think
[13:19] <Vaizki> running esxi free version with a few VMs for my private projects.. and this IRC bouncer :)
[13:19] <AndyEsser> I used to run some IBM eServer Xeon machines at home
[13:19] <Vaizki> haha
[13:19] <AndyEsser> but now I just do everything on amazon instead
[13:19] <Vaizki> eserver.. those are noisy and power hungry
[13:19] <AndyEsser> ha yes
[13:19] <AndyEsser> except for a linux 'server' that I have at home, but currently doesn't do anything yet
[13:20] <Vaizki> I have one of those, it draws about 430 watts.. idle.
[13:20] <AndyEsser> used to have my Jenkins CI on but doesn't anymore
[13:20] <fsphil> I have a little HP microserver. it's rather cute
[13:20] <AndyEsser> <3 HP servers
[13:20] <AndyEsser> :)
[13:20] <Vaizki> there will be some great small servers out with Xeon-D
[13:20] <Vaizki> supermicro already put out a mobo though it's not cheap
[13:21] <Vaizki> http://www.tweaktown.com/reviews/7171/supermicro-x10sdv-tln4f-intel-xeon-server-motherboard-review/index.html
[13:21] <Vaizki> has dual 10G ethernet on the SoC which is nice
[13:22] <AndyEsser> nice
[13:22] <chris_99> has 10G ethernet come down in price, last time i looked i thought cards for that where like £200+
[13:23] <AndyEsser> XeonD-HAB :)
[13:23] <chris_99> heh
[13:24] <Vaizki> the 10G prices have not come down that fast but it's accelerating
[13:24] <chris_99> ah cool
[13:25] <Vaizki> there is a bit of a pain with cat6 10G because it's not available as SFP+ modules
[13:25] <Vaizki> I mean 10GBaseT or whatever it's called
[13:25] <chris_99> SFP+ means what sorry?7
[13:25] <AndyEsser> little modules you plug into SFP ports on switches
[13:25] <Vaizki> small formfactor pluggable module for 10G .. :)
[13:25] <AndyEsser> s/switches/devices
[13:25] <chris_99> ah
[13:26] <Vaizki> if you remember AUI transceivers.. ? :)
[13:26] <Vaizki> (maybe I'm too old)
[13:27] <Vaizki> anyway, it's a common plugin module spec for 1G (SFP) and 10G (SFP+) where you put in the module for the physical transport you're using (copper, single mode fiber etc)
[13:27] <Vaizki> but SFP+ doesn't provide enough voltage/current to run 10G over cat6 copper so those modules do not and won't exist
[13:27] <chris_99> ah hmm, what do you call a cat6 ethernet card then, i just ebay'd some and they seem to be the SFP+ ones
[13:28] <AndyEsser> tbh I'm not sure Id want to run 10g over ethernet
[13:28] <fsphil> I'm currently all wifi, 56mbit/s :)
[13:28] <AndyEsser> eww
[13:28] <fsphil> it's a bit annoying
[13:29] <Vaizki> AndyEsser, why not? my home NAS box is severely limited by 1G ethernet
[13:29] <Vaizki> chris_99, umm? link? :)
[13:29] <AndyEsser> Vaizki: what on earth are you doing with it?! :P
[13:29] <Vaizki> anyway, I see 10G over copper making headway this year and prices coming down
[13:29] <chris_99> Vaizki, http://www.ebay.co.uk/itm/HP-516937-B21-Dual-port-10Gb-Ethernet-Card-518001-001-Mellanox-ConnectX-2-SFP-/321941343106
[13:30] <Vaizki> AndyEsser, it has for example all my raw images from photography
[13:30] <AndyEsser> Vaizki: if you need the constant througput - run fibre :)
[13:30] <Vaizki> chris_99, yea so that has 2x SFP+ cages so you need to add modules to use it
[13:31] <chris_99> Vaizki, mmm, what i was wondering is what do you call cards that have cat6 connectors on?
[13:31] <Vaizki> AndyEsser, well my imac is a little low on fiber or pci-e to stick things into :)
[13:31] <AndyEsser> RJ45
[13:31] <AndyEsser> Vaizki: but it has a 10G card?
[13:31] <Vaizki> no it doesn't.. yet
[13:31] <Vaizki> http://www.ebay.co.uk/itm/Intel-chipset-X540-T2-10G-Dual-RJ45-Ports-PCI-Express-Ethernet-Network-Adapter-/111636496956?hash=item19fe0dc23c:g:k8sAAOSw34FVHSPQ
[13:31] <chris_99> ah cheers Vaizki, heh the price seems to jump a lot
[13:33] <Vaizki> I think the cheapest switches are still netgear.. but some others will join the fray.
[13:33] <Vaizki> they are about £100/port I would guess
[13:35] <mfa298> chris_99: if you want 10G Copper stuff look at something like the X520-T1 (also 540 and T2 variants)
[13:35] <mfa298> although for lots of servers on 10G SFP+ and DA (Direct attach) cables are usually better (cheaper and lower latency)
[13:37] <Vaizki> but copper 10G is coming on motherboards, we already have a bunch of servers with dual-10G on them standard
[13:37] <mfa298> but 10G over cat6 is limited to something like 30m, you really want 6A or 7 for 10G
[13:37] <Vaizki> right, well mostly we use it within the same rack
[13:38] <Vaizki> and DA cables.. well let's just say that they are sometimes compatible between brands :D
[13:38] <mfa298> For same rack SFP+ and DA is the way to go unless you want to ruin your rack with netgear stuff.
[13:39] hb9fdk (bc9b23cd@gateway/web/freenode/ip.188.155.35.205) joined #highaltitude.
[13:39] <Vaizki> well we don't have netgear stuff
[13:39] <mfa298> the only times I've had issues with DA is using them for 1G where some manufacturers (HP) are evil and the same issue applies with non HP branded optic modules.
[13:40] <mfa298> the only other issue is some manufacturers (Brocade) need active DA cables rather than Passive ones, but we've used non Brocade Active DA cables without an issue
[13:40] <Vaizki> yea well we're so cheap we have hp everywhere :(
[13:41] <Vaizki> but some of the HPs we have include both SFP+ and RJ45 10G
[13:41] <Vaizki> and they are cheap as chips
[13:41] <mfa298> I know on their procurve switches they used to accept non HP SFP modules, then disabled it in a software update,
[13:41] <Vaizki> but this is really just R&D lab use, we don't run any production services here
[13:42] <fsphil> now that's evil
[13:42] <mfa298> I've used a bunch of HP switches as well for 10/100 and 1G
[13:42] <Vaizki> fsphil, yes kinda.. but TBH they always claimed only HP modules will work
[13:42] <Vaizki> but non-HP ones worked contrary to documentation and specs
[13:43] <fsphil> still evil. fine if they don't support non-HP configs but they should leave it up to the user
[13:43] <mfa298> most other network switches I've seen just spam the logs for non branded SFP/SFP+ modules or don't provide as much information.
[13:44] <hb9fdk> Hey waizki/fsphil, sorry I did not get the start of your thread. Are you using wlan hw for HABing?
[13:44] <mfa298> I've even see SFP modules branded by a supplier that the suppliers switch then states isn't a supported module (wasnt on HP for that one)
[13:45] <fsphil> hb9fdk: no, though I'm going to use wifi to send video of the launch from the hab perspective
[13:45] <AndyEsser> fsphil: given that HP support is pretty awesome, I'm happy to pay a bit more for HP modules :)
[13:45] <Vaizki> 10G single-mode wlan? no :D
[13:46] <hb9fdk> OK thanks.
[13:46] <Vaizki> AndyEsser, the issue is when you have 2 switches like me.. one is an embedded telco system with SFP+, the other is a HP with SFP+ .. now you can't put a DAC cable between them because the passive cables expose the broadcom to HP so it refuses to work
[13:47] <Vaizki> so I have to put in SFP+ modules + fiber for a 70cm run
[13:47] <AndyEsser> heh
[13:47] <AndyEsser> I wish fibre was cheap enough to do away with all ethernet here :)
[13:47] <Vaizki> fibre is cheap, transceivers aren't
[13:47] <AndyEsser> s/fibre/a fibre infastructure
[13:47] <AndyEsser> :P
[13:48] <mfa298> 1G stuff is fairly cheap on ebay these days, but the 10G stuff is a bit pricey still
[13:48] <Vaizki> I just hope 10GBaseT makes a mass market entry to motherboards, it will drive down prices across the industry, even enterprise stuff
[13:48] <Vaizki> a rising tide and all that
[13:49] <Vaizki> and did you guys see that new pci-e/ethernet chip intel announced.. errr whatsitsname..
[13:49] <AndyEsser> whatsitsname is a great product name :)
[13:50] <Vaizki> it had a realllly large number in the name
[13:51] <Vaizki> FM10000!
[13:52] <AndyEsser> nice
[13:52] <AndyEsser> http://www.intel.com/content/www/us/en/embedded/products/networking/ethernet-multi-host-controller-fm10000-family-overview.html
[13:52] <Vaizki> 36 x 10G from one bridge
[13:52] <Vaizki> or 6x 100G
[13:52] <AndyEsser> Supported Operating Systems: Linux
[13:52] <AndyEsser> :)
[13:53] <chris_99> mfa298, sorry was afk, i'll investigate that thanks
[14:01] trn (jhj@trnsz.com) left irc: Remote host closed the connection
[14:01] <fab4space> daveake, Upu and all, http://cpham.perso.univ-pau.fr/LORA/resources/RPIgateway.pdf it talks about lora and another gateway they have made with a sx1276 module, and they mention habsupplies and piinthesky in the doc
[14:02] <UpuWork> I really need to sort my product images out
[14:11] trn (jhj@trnsz.com) joined #highaltitude.
[14:12] <fsphil> one of those cheap ebay kits are good for small products like that
[14:14] <Vaizki> you mean a light tent / box / cube?
[14:14] <fsphil> yeah
[14:16] <UpuWork> yeah I Have one
[14:16] <UpuWork> just the camera on the Z3 is pants
[14:16] <UpuWork> Got a little Canon now for it
[14:16] <AndyEsser> Z3 isn't _that_ bad
[14:16] <UpuWork> it is
[14:17] <UpuWork> its rubbish
[14:17] <AndyEsser> as in.. the Xperia Z3?
[14:17] <UpuWork> yes
[14:17] <UpuWork> ok try this
[14:17] <UpuWork> imagine you're next to someone who is asleep, its a quiet room. Your phone is on noisy. Click the button a few times to reduce the volume
[14:17] <UpuWork> whats the last thing it does before it goes on silent ?
[14:17] <UpuWork> Lets the loudest beep out it can
[14:18] <AndyEsser> haha
[14:18] <AndyEsser> my phone is always on silent anyway
[14:18] <UpuWork> I like the phone
[14:18] <fsphil> that's very H2G2
[14:18] <UpuWork> but the camera is pants and the interface is years behind Apple
[14:18] <AndyEsser> I had the Z1 and now the Z3
[14:18] <fsphil> it's ok, apple are fixing that by screwing up the iphone
[14:18] <AndyEsser> both I've broken in the exact same way
[14:18] <UpuWork> go change your password on an Exchange account. You have 45s to find where to do it
[14:19] <AndyEsser> and still another year before I can renew contract again :(
[14:19] <AndyEsser> UpuWork: on a phone?
[14:19] <AndyEsser> I wouldn't
[14:19] <UpuWork> on the Z3
[14:19] <UpuWork> and my main issue is Sony install so much crap on there you can't find anything (or remove the crap)
[14:19] <UpuWork> but I still like it :)
[14:19] <AndyEsser> yea, I hate shiteware
[14:19] <AndyEsser> same with laptops (not just sony) as well
[14:20] <fsphil> probably best to avoid sony these days
[14:22] <AndyEsser> I should've gotten the Z3 compact though
[14:22] Action: AndyEsser grumbles and stupid small hands
[14:22] <AndyEsser> at*
[14:23] <fsphil> do they still insist on their own memory card format?
[14:23] gaz (577fc485@gateway/web/freenode/ip.87.127.196.133) joined #highaltitude.
[14:23] <mfa298> at least on a PC you can grab an clean OS image and install without bloatware, not so easy to do that on a phone.
[14:23] Nick change: gaz -> Guest66853
[14:24] <fsphil> sadly no longer true, for windows at least
[14:25] <fsphil> the bios has the ability to install stuff in windows these days
[14:25] Guest66853 (577fc485@gateway/web/freenode/ip.87.127.196.133) left irc: Client Quit
[14:26] <fsphil> http://www.redmondpie.com/lenovo-pcs-can-reinstall-bloatware-even-if-you-format-and-clean-install-windows-but-theres-a-fix/
[14:28] <mfa298> sounds like a good reason not to buy Lenovo (at least until the others start doing the same)
[14:28] <fsphil> it's a windows 'feature' :/
[14:28] <fsphil> been there since win8 apparently
[14:29] <fsphil> they infected win7 through other methods
[14:29] <fsphil> overwriting autochk.exe
[14:33] trn (jhj@trnsz.com) left irc: Excess Flood
[14:33] trn (jhj@trnsz.com) joined #highaltitude.
[14:39] <Vaizki> I would not run windows on bare metal at all but using for example Airspy under vmware just doesn't work well enough.. and yes I have VT-d I/O virtualization and all that jazz :)
[14:40] <fsphil> shame
[14:40] <eroomde> been there with solidworks
[14:40] <eroomde> alrightish for v small things but it just dies very quickly when this get more complicated
[14:41] <Vaizki> and I used ESXi hypervisor on a Xeon E3 where I mapped the PCI-e graphics card directly through to win7
[14:41] <Vaizki> so graphics performance was very good (native really)
[14:41] <chris_99> hmm there doesn't seem to be a simple way to tell if you've got vt-d
[14:41] <AndyEsser> Yea, trying to do direct hardware IO is terribad on any virtualisation platform sadly
[14:41] <chris_99> (via linux)
[14:41] <Vaizki> chris_99, just check the intel ark pages
[14:42] <Vaizki> AndyEsser, well VT-d works for pci-e devices so maybe if I got a pci-e usb expansion card it could be done.. not sure why it's not allowing usb hub chips on the motherboard to be mapped through
[14:42] <chris_99> awh i don't have it :'(
[14:43] <Vaizki> maybe KVM could do it
[14:43] <AndyEsser> Vaizki: don't know if it's maybe something to do with memory-mapped IO or something on the host
[14:45] <mfa298> I think the vt-d stuff also needs some support in the hardware being mapped/shared.
[14:45] <Vaizki> well.. I dunno for sure but I think VT-d really enables native access including memory mapped I/O.. and the chipset/cpu handle the translations
[14:46] <AndyEsser> Yea, I'd have thought VT-d would do it
[14:46] <AndyEsser> sad times :(
[14:46] <Vaizki> IOMMU is for DMA/memorymap like MMU is for RAM
[14:47] <mfa298> I think originally it was about making it easier to share bits of hardware rather than having to emulate everything, but Ive not looked at it in detail for a couple of years. Originally it was just suport in the CPU related to ring0
[14:48] <mfa298> but that was the original Intel-VT (and AMD had their version that was a bit better), then Intel had VT-d which was better again
[14:48] <Vaizki> yea amd has IOV or something
[14:50] <fab4space> I have the xperia Z3 compact, its the best phone I have ever owned
[14:51] <fab4space> AndyEsser,
[14:52] <AndyEsser> :)
[14:52] <AndyEsser> when my contract is up for renewal, I'll likely go for the Z5 compact or whatever generation it is at the time
[14:53] <eroomde> i need a new phone
[14:54] <eroomde> this one is only 3 yeras old but that seems to be about the lifespan of a smartphone
[14:54] <AndyEsser> tbh... I could probably easily get buy with a phone that just does SMS and Calls
[14:54] <eroomde> it really creaks
[14:54] <eroomde> i'd love to do that
[14:54] <eroomde> srsly
[14:54] <AndyEsser> for a geeky/tech guy, my "smart" phone is highly under utilised
[14:54] <eroomde> a nokia 105 and i'd be happy
[14:55] <eroomde> but, i really do like having a browser and maps for travel
[14:55] <AndyEsser> I use WhatsApp for messaging (which I'd miss), and I have FB/Twitter - but I have those on my PC all the time anyway
[14:55] <AndyEsser> eroomde: yea, I use mine as a SatNav when in the car, or desperately trying to find somewhere when in a town
[14:55] <eroomde> for maps of foreign cities, public transport timetables and so on, it's invaluable
[14:56] <AndyEsser> Like... I'd love to write some apps for my phone, and possibly hook them into my Microsoft Band
[14:56] <AndyEsser> but developing for Android is... god awful
[14:56] <gonzo_> do you have the collection of pistit notes stuck to your smart phone, using the 'memo' functionality?
[14:56] <gonzo_> post-it
[14:56] <AndyEsser> no, that'd be the various notebooks I have around the house and office and can never remember which one I've written stuff in :)
[14:57] <gonzo_> random access mem then
[14:57] <eroomde> i wondered about maybe having a nokia 105 and an ipad air with a data plan or something for travelling
[14:57] <eroomde> but then that's 2 things
[14:57] <fsphil> my brain is definitly random access
[14:58] <AndyEsser> heh
[14:58] <fsphil> the idea behind smartphones is fine, they just need better battery life
[14:58] <gonzo_> I have a piece of perspex sheet on my desk, with a sea of post its. So I often walk off with whole rafts of to-do's stuck to the underside of the forearms
[14:59] <gonzo_> I tried the ultra low power mode on my s5 a while ago. Was amazed it actually lasts
[14:59] <gonzo_> just enough to get text/calls
[14:59] <gonzo_> which is mostly what it needs to do at idle
[14:59] <AndyEsser> gonzo_: I used to have in my lounge wall post-its all over one wall
[14:59] <eroomde> the nokia 105 is <20 and has over a month on standby
[14:59] <AndyEsser> landlord wasn't impressed when he came around
[14:59] <eroomde> <£20*
[15:00] <fab4space> with z3 compact the battery life is very good when using all the features, 2 two days without charging
[15:00] <gonzo_> £ standby?
[15:01] <eroomde> ?
[15:01] <gonzo_> it needs an idle time of a week between charges to be really useful
[15:01] <gonzo_> soty ed, I read the last as <£20 on standby
[15:02] <adamgreig> my watch gets about a week between charges.. ;)
[15:02] <gonzo_> sorry
[15:02] <AndyEsser> I should really take my 'test battery' out of the freezer...
[15:02] <AndyEsser> I have to charge my Band every other night... which is crap for monitoring sleep
[15:02] <adamgreig> mm
[15:02] <adamgreig> pebble does health tracking and sleep stuff now
[15:02] <gonzo_> why do freezers not have the little symbols on for the length of time you can store batts in
[15:02] <AndyEsser> also annoyed that they brought out the Band 2 just 6 months after the Band
[15:03] <AndyEsser> gonzo_: hehe
[15:03] <gonzo_> 50mhz?
[15:03] <eroomde> what is a band?
[15:03] <AndyEsser> Microsoft Band
[15:03] <AndyEsser> smartwatch
[15:03] <AndyEsser> https://www.microsoft.com/microsoft-band/en-gb
[15:03] <fsphil> :/
[15:03] <eroomde> why do you monitor your sleep?
[15:03] <gonzo_> Sorry, band1 is 50MHz
[15:03] <AndyEsser> eroomde: because I haven't had a good night's sleep in about 6-7 years
[15:04] <adamgreig> i don't any more but for a couple years i tracked how much sleep and when every night
[15:04] <eroomde> does the band improve your sleep?
[15:04] <adamgreig> it was quite fun, can see patterns
[15:04] <gonzo_> worrying about your watch?
[15:04] <AndyEsser> eroomde: no, but it can help me narrow down if there are things I'm doing to affect my sleep
[15:04] <AndyEsser> and it also gives a rough idea of the quality of sleep I got and things
[15:04] <eroomde> i'm going to sound like a graumpy old man
[15:05] <eroomde> but a lot of this quantified self stuff seems to pretend to be about fixing but is actually just measuring
[15:05] <AndyEsser> agreed
[15:05] <AndyEsser> but I'm never going to turn down more data
[15:05] <AndyEsser> :)
[15:06] <eroomde> i'd bet £100 than not wearing a band, but concentrating on eating better, exercising more, and going to bed such that you in theory can get 7-8hours of sleep before having to get up to work the next day, would improve things
[15:06] <eroomde> much more than merely having it digitally confirmed that you're not behaving well
[15:07] <eroomde> i had a colleague (keeping dates vague) who had loads of trouble sleeping and kept buying all these monitors and stuff to 'try and home in on it'
[15:07] <AndyEsser> I also like having something merely 2 inches away from my phone that notifies me of things :)
[15:07] <eroomde> the rest of us were all thinking, but not saying, 'YOU'RE TWENTY STONE. WHAT DO YOU EXPECT'
[15:08] <AndyEsser> I'm not 20 stone ;)
[15:08] <eroomde> eventually after about 8 months they came into work late having finally been to see a doctor and acted like some great mystery had finally been revealed by an oracle - they had sleep apnea
[15:08] <eroomde> from being too fat
[15:08] <AndyEsser> heh
[15:09] <eroomde> i think they probably knew really, but buying all this stuff and pretending it was a radon leak from the radiator was just displacement activity to put off the inevitable
[15:09] <eroomde> lifestyle change. no one else to blame it on. sucks.
[15:10] Laurenceb (~Laurence@host86-179-250-250.range86-179.btcentralplus.com) joined #highaltitude.
[15:11] <eroomde> equally, if having something quantified helps you do what you could have done anyway, it helps right
[15:12] <eroomde> i'm sure my fitness pal has helped people achieve the same thing as telling people to just eat less generally and more veg as a proportion
[15:12] <AndyEsser> some people react better to being able to visualise the stats etc
[15:12] <AndyEsser> it's actually quite difficult to visually see weight loss in yourself
[15:12] <AndyEsser> at my last job I used to cycle to/from the office
[15:12] <AndyEsser> about 4 miles each way
[15:13] <AndyEsser> I used Strava to record the journies because I could then see myself improving, and then also got competitive with myself
[15:13] <Vaizki> for me measuring things makes goals easier to reach
[15:13] <AndyEsser> "I will do the journey in less than 17 minutes, I will hit this speed at this point, etc"
[15:13] <AndyEsser> journies?... that word doesn't look right
[15:14] <fsphil> sounds like mario, only less fun
[15:14] <fsphil> mario ghost racing*
[15:14] <AndyEsser> fsphil: well I guess that's the concept behind the "Zombies! Run!" running app
[15:14] <AndyEsser> "gamification"
[15:15] <Laurenceb> SaaS
[15:15] <eroomde> https://twitter.com/FinalBullet/status/676379766308630528
[15:15] <AndyEsser> heh
[15:15] <AndyEsser> Laurenceb: I have all manner of SaaS...o wait, that's Sass... my bad :P
[15:16] <eroomde> one thing i would v much like actually is a 4 mile commute
[15:16] <eroomde> something cyclable easily and walkable if i had some time and fancied a walk
[15:17] <eroomde> atm is 23 miles
[15:17] <AndyEsser> eroomde: I had to cycle in from Landbeach to Cambridge Science Park
[15:17] <eroomde> i had a period of cycling it in summer but i'm just very, very tired by the end of each week
[15:17] <AndyEsser> but I didn't want to go near the A road
[15:17] <AndyEsser> thankfully nice convenient path that ran though fields all the way to the bac kof the science park
[15:17] <eroomde> and when it gets darker earlier in the evenings and you've had a trying day and have a 23 mile cycle ahead of you you're seriously tempted to just sleep under the desk at work
[15:17] <AndyEsser> basically off-road, and got very muddy in the winter - but fun nonetheless
[15:18] <AndyEsser> eroomde: yea, I wouldn't be able to do that
[15:18] <AndyEsser> my commute now is 10 minute walk :)
[15:18] <fsphil> I could probably do office to home in 20 minutes
[15:19] <fsphil> I really should do it
[15:19] <fsphil> all this sitting around is no good
[15:19] <Laurenceb> I could laugh at gamification, but atm I making gantt charts :-/
[15:19] <AndyEsser> despite the walking to/from office, I should still go out for a 30 minute walk each evening, but I'm just useless like that
[15:19] <AndyEsser> need to get on my exercise bike more
[15:19] <AndyEsser> 1 or 2 times a week isn't enough
[15:20] <fsphil> you smoke :p
[15:20] <AndyEsser> I fail to see your point
[15:20] <AndyEsser> :P
[15:20] <Vaizki> a bit OT but anyone used Ultracell batteries? need 12V/7Ah for my UPS and Ultracell are on sale...
[15:20] <eroomde> Laurenceb: i'm not sure if i would call gantt charts gamification
[15:20] <AndyEsser> Vaizki: nice way to swing us back on topic :)
[15:20] <eroomde> perhaps, though, the sport of fools
[15:20] <Laurenceb> same level of BS most of the time
[15:20] <fsphil> would be like me walking each day, but to mcdonalds :)
[15:20] <Laurenceb> heh
[15:21] <AndyEsser> heh
[15:27] <eroomde> there exists the phrase 'a plan is useless, but planning is useful'
[15:27] <eroomde> i agree with this
[15:28] <eroomde> i think, to a first approximation, funding bodies agree with this in their hearts
[15:28] <eroomde> they just want to see you've thought about it
[15:28] <eroomde> i resent, then, that the plan must involve so much actual presentation work
[15:28] <eroomde> like, for example, the priduction of gantt charts
[15:28] <eroomde> production*
[15:29] <AndyEsser> eroomde: how would you know how much funding you needed, if you don't take into account the cost of labour, and how can you calculate that without planning out the thing?
[15:29] <AndyEsser> (playing devils advocate btw)
[15:30] <eroomde> because like all funding applications ever you greatly exaggerate hours
[15:30] <adamgreig> and you can't hope to accurately plan how much time it will take anyway
[15:30] <adamgreig> not saying you don't try, but..
[15:30] <eroomde> you hope 2 of you can do it in 3 months spending about 50% of your time on it
[15:30] <eroomde> so you put 3 people down full time for 6 months
[15:31] <eroomde> and that's fine because the sort of people they fund are used to that
[15:31] <AndyEsser> fair enough
[16:04] <DoYouKnow> hi
[16:05] <DoYouKnow> I have instruction for commanding the UBLOX7 gps dongle to generate an fft and send it over the usb port here: http://pastebin.com/tzmFDgsr
[16:16] <Laurenceb> what on earth
[16:17] <Laurenceb> an fft of what?
[16:26] ibanezmatt13 (~ibanezmat@host81-131-186-143.range81-131.btcentralplus.com) left irc: Ping timeout: 240 seconds
[16:28] <Lunar_Lander> done https://drive.google.com/file/d/0B1lke6hPHKJDZzdXaThrMmltSUE/view
[16:29] <fsphil> your own little cubesat
[16:30] <AndyEsser> fsphil: might need a little nudge from a HAB though :P
[16:31] <Lunar_Lander> :)
[16:31] <fsphil> ROCKETS!!!1
[16:32] <fsphil> most rockets might not get as high as a balloon, but they do it in style
[16:33] <AndyEsser> fsphil: I want to launch a rocket from a balloon....
[16:33] <AndyEsser> I believe eroomde called it a Rockoon?
[16:33] <fsphil> yeah
[16:33] <fsphil> Ballocket
[16:38] <Lunar_Lander> :)
[16:39] <nick_> No, a rockoon is a racoon you taught to play Iron Maiden songs...
[16:39] <AndyEsser> Death Metal Racoon
[16:40] <AndyEsser> http://i.kinja-img.com/gawker-media/image/upload/zmrsvf3c7hcqxjzhehen.jpg
[16:40] <nick_> Racoons have tiny people hands. I'm sure you could make them a little guitar.
[16:41] <bradfirj> Lunar_Lander: Looks great, what radio module are you planning to use?
[16:41] <bradfirj> you've got lots of space, wondering what the choice is in that case
[16:41] <Lunar_Lander> thanks :)
[16:41] <Lunar_Lander> MTX2
[16:42] <adamgreig> are you worried about putting the MTX2 right next to the GPS receiver?
[16:43] <eroomde> the socratic method in action.
[16:44] <AndyEsser> u wot m8
[16:45] <bradfirj> The harmonics from an ISM band radio shouldn't get near GPS frequencies, right?
[16:46] <fsphil> the old ntx2 (pre-b) put out enough energy at ghz frequencies to be decoded at the other side of the room
[16:46] <eroomde> intermodulation
[16:46] <eroomde> when they're that close it'll just get injected straight in after any filters
[16:46] <Lunar_Lander> damn
[16:46] <bradfirj> Software interlocks :)
[16:46] <fsphil> the mtx2 might be a lot better
[16:47] <eroomde> tbh i think you'll get away with it
[16:47] <bradfirj> Shut down the transmitter before taking a gps reading if you're worried
[16:47] <mattbrejza> its probably fine Lunar_Lander , many people have done muich worse
[16:47] <Lunar_Lander> I thought it is OK due to this https://www.flickr.com/photos/randomskk/14728883460/in/album-72157645881007740/
[16:47] <fsphil> lol indeed
[16:47] <mattbrejza> looks about the same as the pits
[16:47] <adamgreig> haha yes I was about to link to that
[16:47] <adamgreig> it'l probably be fine indeed
[16:47] <adamgreig> the mtx2 is better
[16:48] <adamgreig> but when you have like five PCBs and as much space as you might want, sticking the most powerful source of RFI next to the most sensitive receiver of RFI is maybe not the best possible of all layouts
[16:48] <Lunar_Lander> yea
[16:48] <Lunar_Lander> I agree
[16:49] <Lunar_Lander> so the next revision could address that issue I think
[16:51] <Lunar_Lander> the good thing is, you could stack protoboards inbetween
[16:55] <AndyEsser> I plan on putting my NTX2 and GPS modules diagonally opposite corners of each other, just to be on the safe side
[16:55] <adamgreig> NTX2 or NTX2b?
[16:55] <Lunar_Lander> sounds good
[16:55] <AndyEsser> NTX2b
[16:56] <adamgreig> good good
[16:56] <AndyEsser> GPS will be the next thing I purchase and get working, and then the NTX2b
[16:56] <AndyEsser> then I should have everything in and working that I need for my next launch
[16:56] <AndyEsser> Ideally hope to order the GPS soon (TM) but waiting on clients... gah
[16:56] <AndyEsser> next launch? err... first launch
[16:57] <Lunar_Lander> I got a naive question
[16:58] <Lunar_Lander> does the old method with the resistor network from the Wiki still work on the MTX2
[16:58] <Lunar_Lander> ?
[16:58] <fsphil> yes
[16:58] <fsphil> it is still an analogue input
[16:58] <Lunar_Lander> thanks
[16:58] <AndyEsser> resistor network?
[16:58] <Lunar_Lander> yea
[16:59] <Lunar_Lander> https://ukhas.org.uk/guides:linkingarduinotontx2?rev=1368727224
[16:59] <AndyEsser> Danke
[17:01] <adamgreig> aww, my "detailed explanation" image is gone from the latest version
[17:02] <Lunar_Lander> yeah that was awesome
[17:02] <Lunar_Lander> bitte andy
[17:02] <AndyEsser> Only ever use two frequencies with RTTY?
[17:03] <adamgreig> rtty only uses two frequencies yes
[17:03] <fsphil> it's basically rs232, only with frequency shift rather than voltage
[17:04] <AndyEsser> Yea, I guessed, just thought maybe using 8 frequencies or something and having symbols rather than 1's and 0's might allow for a better transmission rate
[17:04] <adamgreig> you get better spectral efficiency but it's not necessarily "better", depends on things
[17:04] <adamgreig> in our case though yes - it would be better
[17:05] <adamgreig> some people use dominoex and contestia which are mFSK (vs rtty is 2FSK/BFSK)
[17:05] <adamgreig> a lot of people using lora these days which is a different thing again
[17:05] <AndyEsser> might be things for me to look at in the future
[17:05] <adamgreig> but rtty is really simple and works fine for simple telemetry strings, so
[17:05] <AndyEsser> will stick to plain old 2-tone RTTY to start with :)
[17:06] Action: AndyEsser hides before eroomde can shout at him
[17:07] <gonzo_> simple rtty works out to the horizon, where the signal falls off a cliff anyway
[17:08] <bradfirj> Do other modes perform OTH?
[17:08] <adamgreig> sure but you could get a _lot_ more data over that link out to the horizon than rtty would allow
[17:08] <adamgreig> not really - you might get a bit further down the cliff
[17:09] <adamgreig> lora & contestia will decode when you can't hear or see anything yourself, but it just gives you a littl emore
[17:09] <bradfirj> Or we go down the well trodden route of flying a car battery, that way you can bounce off the ionosphere :p
[17:09] <adamgreig> you can do HF with a 10mW transmitter and also bounce off the ionosphere
[17:09] <adamgreig> no need for a car battery
[17:10] <adamgreig> you have other serious problems though, like antennas
[17:10] <adamgreig> and reception stations, skip distances, propagation conditions, bla bla
[17:10] <bradfirj> All in all, 'effort'
[17:10] <adamgreig> the antenna might go beyond 'effort' too
[17:10] <adamgreig> very long piece of conductive metal wire danging from a balloon
[17:10] <bradfirj> What does a 'small' HF antenna look like?
[17:11] <bradfirj> Are we talking 3ft or 3m
[17:11] <bradfirj> Or 30
[17:11] <adamgreig> depends on what kind of hf but a dipole for say 20m is 20m long
[17:11] <adamgreig> so..
[17:11] <bradfirj> Crumbs
[17:11] <adamgreig> there are other antenna geometries which are more compact but generally less efficient or have weird radiation patterns
[17:11] <adamgreig> neither of which is likely to work in your favour when operating at very very low powers from a spinning balloon
[17:12] <russss> or are RX-only
[17:12] <adamgreig> yea
[17:12] <adamgreig> (in case it wasn't clear, the long conductive antenna is how you start forest fires)
[17:12] <bradfirj> That's why you need cold gas thrusters and an attitude control system
[17:12] <bradfirj> Sheesh, amateurs
[17:12] <bradfirj> :)
[17:13] <adamgreig> people have done it though (hf from a balloon)
[17:13] <adamgreig> worked quite well
[17:13] <adamgreig> even on pico flights / long duration floaters
[17:13] <adamgreig> you can also do the satellite thing instead
[17:13] <adamgreig> or aprs and rely on an extant network of receivers..
[17:13] <bradfirj> Was the forest fire inducing antenna required or did they get away with 1/2 or 1/4 dipole
[17:14] <bradfirj> Yeah I've used rockblocks for ground based stations, remote monitoring and so on, but in comparison to a pico I may as well have 230V mains at those sites
[17:14] <bradfirj> The rockblocks aren't exactly power efficient
[17:15] <adamgreig> or lightweight
[17:15] <adamgreig> iridium probably not the way to go
[17:16] <fsphil> good choice for ocean buoys
[17:17] <bradfirj> They work very well for Rallying
[17:17] <bradfirj> Where we have fixed stations in the middle of nowhere
[17:17] <fsphil> yeah sounds ideal for that
[17:18] <bradfirj> Except when the idiot in charge of the point keeps picking the transmitter up to show it to people
[17:18] <bradfirj> >.>
[17:18] <bradfirj> Needs a big label WARNING SATELLITE RADIATION
[17:18] <fsphil> you didn't tell them it was radioactive?!
[17:18] <bradfirj> I mean it wouldn't be lying
[17:19] <bradfirj> adamgreig: What other satellite networks are in a hobbyist's reach?
[17:19] fab4space (~fab4space@109.237.242.98) left irc: Ping timeout: 256 seconds
[17:19] <AndyEsser> lets make our own :)
[17:19] <bradfirj> I mean it wouldn't be impossible....
[17:19] <adamgreig> nothing obviously good. I think some people here were looking at ARGOS but it's not generally available
[17:22] <bradfirj> Even if you break them out of the rockblock board, the iridium 9602 is large and heavy so you're right about that one sadly
[17:27] <Laurenceb> ARGOS is ideal, but really hard to get an upload token
[17:28] <AndyEsser> http://www.argos.co.uk/? :P
[17:28] <Laurenceb> lol
[17:28] <Laurenceb> you also need to get your hardware approved, but that doesnt actually look too hard
[17:28] <Laurenceb> there is a lab in france that will certify your kit for free if you ask really nicely
[17:29] <Laurenceb> you need to do that before you have any chance of getting tokens
[17:29] <AndyEsser> "Please sir, can I have certification?"
[17:29] Action: AndyEsser sends package with cheese and wine
[17:29] <adamgreig> don't try sending cheese and wine to the french :P
[17:29] <AndyEsser> adamgreig: it would be french wine and cheese... not english stuff
[17:29] <adamgreig> hehe
[17:30] <Laurenceb> you'd need to be part of a university and explain why it was scientifically useful
[17:30] Action: AndyEsser goes and "borrows" some University of Chester business cards
[17:30] <Laurenceb> but aiui the system seems to be rather congested
[17:30] <AndyEsser> https://camo.producthunt.com/e5d034b5c25acd9bf5bf580620803436d15e099d/687474703a2f2f6e69636b74756d6d696e656c6c6f2e636f6d2f77702d636f6e74656e742f75706c6f6164732f323031332f31322f426563617573652b6f662b736369656e63652b5f37616364373138346632383161333834633136636138653264666562366336312e706e67
[17:30] <Laurenceb> people are sticking ARGOS on ten thousand turtles and stuff
[17:30] <AndyEsser> large print out of this as well
[17:30] <AndyEsser> wow
[17:31] <Laurenceb> so it seems to be very hard to get the 32 bit tokens you need to upload and grab your data from their server
[17:32] <Laurenceb> established research groups have first priority, so the best bet would be to find a group who got too many
[17:32] <Laurenceb> and can resell you some
[17:32] Nick change: michal_f -> michalfAFK
[17:32] <Laurenceb> I looked into this for picoballoons but it was a massive headache and I didnt make any progress :-/
[17:33] <AndyEsser> yea, sounds it
[17:33] <AndyEsser> right time to shut our server down...
[17:33] <AndyEsser> woo
[17:33] <AndyEsser> chat later everyone
[17:33] <AndyEsser> night
[17:35] <Laurenceb> probably have more luck trying to launch a UKHAS picosat :P
[17:35] <Laurenceb> of course if we then implemented an ARGOS style system on it, we'd be bombarded by turtle researchers
[17:36] <Laurenceb> so it'd probably pay for itself :D
[17:37] <AndyEsser> Possibly - but initial capital required would be quite high
[17:38] <adamgreig> how many satellites does argos have?
[17:43] <AndyEsser> *sigh* Hyper-V is so terribad
[17:45] <Laurenceb> not sure
[17:45] <Laurenceb> probably tens
[17:46] <Laurenceb> there is ARGOS on a lot of NOAA weather sats aiui
[17:46] <Laurenceb> https://en.wikipedia.org/wiki/Polar_Operational_Environmental_Satellites
[17:46] <Laurenceb> at least 5 there
[17:47] <Laurenceb> MetOp-A and MetOp-B, then some dedicated ARGOS sats still in service I think
[17:47] <adamgreig> nice
[18:12] hb9fdk (bc9b23cd@gateway/web/freenode/ip.188.155.35.205) left irc: Ping timeout: 252 seconds
[18:17] <AndyEsser> and back
[18:18] <AndyEsser> would you look at that... change my code not to allocate a single 4k array on a 2k device and no more hard resets...
[18:18] <adamgreig> funny that eh
[18:19] <AndyEsser> indeed
[18:19] <AndyEsser> :P
[18:19] <AndyEsser> herp derp
[18:19] <AndyEsser> it's been a while since I was this restricted on device resources - shall try not to make the mistake again
[18:22] <fsphil> it's quite fun working on micros
[18:22] <AndyEsser> indeed
[18:22] <AndyEsser> I love the challenge :)
[18:22] <AndyEsser> need to just wrap up this stuff, then figure out reading/writing to the EEPROM and I'm all set to start playing with the more fun stuff like GPS :)
[18:23] <AndyEsser> only having 1 UART is a bit annoying though
[18:23] <AndyEsser> fsphil: did you say you use the 324 or something similar which as 2?
[18:24] fab4space (~fab4space@AMontpellier-656-1-416-68.w90-0.abo.wanadoo.fr) joined #highaltitude.
[18:29] <fsphil> yea. I had a 644p, which is the same but more ram
[18:30] <AndyEsser> o good, still comes in a DIP package :)
[18:30] <AndyEsser> might have to switch to those
[18:31] <fsphil> larger, but still DIP
[18:32] <fsphil> possibly more moneys too
[18:32] <mattbrejza> its like £7 iirc
[18:33] nickjohnson (sid789@gateway/web/irccloud.com/x-vwpkmcudsdideauq) left irc: Ping timeout: 250 seconds
[18:33] <AndyEsser> 40pin, rather than 28
[18:33] <AndyEsser> but the extra pins will always come in handy
[18:33] <AndyEsser> £2 > £7 is quite the jump
[18:34] <fsphil> http://www.rapidonline.com/electronic-components/atmel-atmega324p-20p-atmega325p-20pu-8-bit-microcontroller-2k-dil-40-73-5080/
[18:34] <fsphil> £5.10
[18:34] <fsphil> still pricey
[18:35] <mattbrejza> ok the 644 is £5 in dip
[18:35] <AndyEsser> the io pins are in a much more logical grouping as well
[18:35] <fsphil> if the 644 is the same price, get that
[18:35] <fsphil> 4k of sram
[18:35] <fsphil> though 2k is a huge amount for a basic tracker
[18:36] <AndyEsser> it won't always be just a basic tracker though
[18:36] <AndyEsser> I'd rather have RAM and resources and not use them
[18:36] <AndyEsser> than have to work on a whole new board when I want to add more sensors or something
[18:36] <fsphil> fair
[18:36] nickjohnson (sid789@gateway/web/irccloud.com/x-barcwregqzkhjbxl) joined #highaltitude.
[18:37] <AndyEsser> £4 on Farnell
[18:38] <fsphil> I might still have a few 644p's in the parts bin
[18:38] <AndyEsser> right, quick re-org of my code, and it's much more stable and easier to read... don't know what I was thinking last night
[18:38] <AndyEsser> it was such a mess
[18:40] <AndyEsser> Don't suppose there's a way to find out how much RAM is in use on the chip?
[18:40] <fsphil> gcc has a size command
[18:40] <adamgreig> you can see how much is statically allocated
[18:41] <adamgreig> avr-size
[18:41] <fsphil> run that on your .out file
[18:41] <adamgreig> but that won't count stack usage
[18:41] <adamgreig> or other runtime dynamic usage
[18:41] <fsphil> my blink program is 260 bytes. what a pig
[18:41] <AndyEsser> Yea, I meant in realtime, as oppose to output
[18:41] <AndyEsser> but checking it in avr-size might be good
[18:42] <adamgreig> in realtime it's quite nebulous!
[18:42] <AndyEsser> fsphil: run that on the output from avr-gcc or avr-obcopy?
[18:42] <AndyEsser> objcopy*
[18:42] <adamgreig> all the ram is "in use" in real time
[18:42] <adamgreig> you don't have an operating system managing this for you
[18:42] <adamgreig> your code might be reading or writing any of it
[18:42] SA6BSS-Mike|2 (~SA6BSS-Mi@81-236-232-232-no30.tbcn.telia.com) joined #highaltitude.
[18:42] <AndyEsser> hmm suppose that's true
[18:42] <fsphil> AndyEsser: probably on the input from the objcopy
[18:42] <fsphil> depends on your build scripts
[18:43] <fsphil> I only use objcopy to convert the .out file to .hex
[18:43] <AndyEsser> 0 2822 0 2822 b06FlightDataRecorder.hex
[18:43] <AndyEsser> fsphil: same
[18:43] <fsphil> ah cool, didn't know -size worked on hex files
[18:43] <adamgreig> AndyEsser: you can look at the current position of the stack pointer to work out stack usage and how close to your highest allocated memory it is
[18:43] <mattbrejza> doesnt tell you anything useful though
[18:44] <fsphil> well, it's different
[18:44] <fsphil> on the hex it dumps everything in the data section
[18:44] <adamgreig> the hex file doesn't tell you about ram usage at all does it?
[18:44] <adamgreig> at that point the information is lost
[18:44] <AndyEsser> 2598 224 591 3413 d55FlightDataRecorder.elf
[18:44] <fsphil> much better
[18:45] es5nhc (~tarmo@108-40-71-217.sta.estpak.ee) left irc: Read error: Connection reset by peer
[18:45] <fsphil> 224 bytes for non-0 static variables
[18:45] SA6BSS-Mike (~SA6BSS-Mi@81-236-232-232-no30.tbcn.telia.com) left irc: Ping timeout: 240 seconds
[18:45] <fsphil> +591 bytes that are zero
[18:45] <fsphil> if I've read that right
[18:46] <adamgreig> 591 bytes of ram that's used byt
[18:46] <adamgreig> but not initialised at declaration
[18:46] <AndyEsser> 224 is the data section, 591 is the bss section
[18:46] <fsphil> it's zero'ed iirc
[18:46] <adamgreig> yea
[18:46] <adamgreig> global and static variables are initialised to all 0s when declared if not otherwise initialised
[18:47] <adamgreig> anything that's allocated to a location in ram
[18:47] <fsphil> 1233 bytes remaining. though some of that will be stack
[18:48] <AndyEsser> fsphil: so just 2k - (bss + data)?
[18:48] <fsphil> should be
[18:49] <fsphil> you can save a lot of ram on the avr by keeping data out of, er, data
[18:49] <AndyEsser> lol
[18:49] <fsphil> if you declare it as PROGMEM it will stay in flash, won't get copied into ram
[18:49] <fsphil> but then you have to use additional commands to read it
[18:50] <fsphil> very handy for strings, avr-libc has some special string functions that can read from flash
[18:50] <fsphil> http://www.nongnu.org/avr-libc/user-manual/pgmspace.html
[18:50] <AndyEsser> well I think the most memory intensive thing is done now (a bunch of stupid string manip)
[18:50] <AndyEsser> everything else should be fairly lightweight
[18:52] <Lunar_Lander> finally back
[18:57] jcoxon (~jcoxon@247.141.112.87.dyn.plus.net) joined #highaltitude.
[19:04] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[19:09] Hix_ (~hix@97e08719.skybroadband.com) left irc: Read error: Connection reset by peer
[19:10] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[19:13] Hix_ (~hix@97e08719.skybroadband.com) left irc: Read error: No route to host
[19:14] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[19:19] Hix_ (~hix@97e08719.skybroadband.com) left irc: Read error: Connection reset by peer
[19:19] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[19:30] Hix_ (~hix@97e08719.skybroadband.com) left irc: Remote host closed the connection
[19:48] geheimnis` (~geheimnis@23.226.237.192) left irc: Remote host closed the connection
[19:49] <AndyEsser> http://imgur.com/VOJN3kS
[19:49] <AndyEsser> :)
[19:50] 17WABB4VS (~geheimnis@23.226.237.192) joined #highaltitude.
[19:50] michal_f (~michal_f@91.146.241.130) joined #highaltitude.
[19:51] jcoxon (~jcoxon@247.141.112.87.dyn.plus.net) left irc: Quit: This computer has gone to sleep
[19:59] <AndyEsser> well that was easy enough to get EEPROM read/write done :)
[20:01] <Ian_> [18:33] AndyEsser, a difference of less than a 20 pack of cigarettes. You deflected the question when asked . . . :-)
[20:02] <AndyEsser> Ian_: hmm... can't see that at all
[20:02] <AndyEsser> what timezone are you in?
[20:04] <AndyEsser> Ian_: also can't find it in the history logs
[20:16] Strykar_ (~wakkawakk@122.170.159.167) joined #highaltitude.
[20:16] <Ian_> [18:33] <AndyEsser> £2 > £7 is quite the jump
[20:16] <Lunar_Lander> hi Ian_
[20:17] <AndyEsser> Ian_: you will be very hard pressed to find a 20 pack of cigarettes for < £5 :P
[20:17] <Ian_> Hi Lunar_Lander, where were they guys when you were asking for comments about your board . . . :)
[20:17] <AndyEsser> But I Get the point ;)
[20:17] <Lunar_Lander> xD
[20:17] <Ian_> No, the difference in the price of the chips was less than . . .
[20:19] Strykar (~wakkawakk@122.170.159.167) left irc: Ping timeout: 256 seconds
[20:19] <Ian_> All times Zulu
[20:21] jcoxon (~jcoxon@247.141.112.87.dyn.plus.net) joined #highaltitude.
[20:22] jcoxon (~jcoxon@247.141.112.87.dyn.plus.net) left irc: Client Quit
[20:23] <Ian_> A good few years ago, I felt the difference between eating at home and having to walk 100m to a mess hall when in barracks for a while.
[20:24] <Ian_> The walk was very beneficial.
[20:30] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[20:35] Hix_ (~hix@97e08719.skybroadband.com) left irc: Ping timeout: 255 seconds
[20:48] <chris_99> has anyone used a pt100 probe here per chance, just looking at http://www.ti.com/product/ADS1248/technicaldocuments
[20:58] <adamgreig> looks nice AndyEsser
[21:01] <Vaizki> I have considered ADS1248 for pt100 but never got around to it
[21:02] <chris_99> cool, this would have a dual purpose for me brewing & HAB
[21:03] <AndyEsser> adamgreig: cheers - the whole Command Line Interface is disabled unless a specific pin is pressed on bootup
[21:03] <Vaizki> The pga makes it great for pt100 so go for it
[21:03] <AndyEsser> s/pressed/high
[21:03] bertrik (~quassel@rockbox/developer/bertrik) joined #highaltitude.
[21:08] <Vaizki> You will need cold junction compensation for hab though
[21:09] <chris_99> oh i read a little about that, does that mean essentially you keep a resistor at say 0C?
[21:11] <chris_99> oh that sounds like that's the ice bath method rather than cold junction compensation
[21:12] <Vaizki> with pt100 you have 2 junctions where the metals meet other metal.. In the prbe tip and the cold junction where the probe hooks up to your board
[21:13] <Vaizki> For accurate measurements you need to measure the temp at the cold junction via another temp probe
[21:15] <chris_99> ah interesting
[21:15] <Vaizki> Tbh I would not bother with thermocouples or pt100 for hab
[21:15] <mattbrejza> pt100 doesnt need any of that
[21:15] <mattbrejza> you just read its resistance
[21:16] <Vaizki> iPhone irc is bad...
[21:16] <craag> yeah you're thinking of thermocouples
[21:16] <craag> pt100 is a resistance thermometer
[21:16] <bradfirj> Vaizki: Colloquy
[21:16] <bradfirj> + ZNC
[21:17] <Vaizki> I have znc and limechat
[21:17] <bradfirj> I built mine with colloquy because there's a znc plugin for push
[21:19] <Laurenceb> http://forum.nasaspaceflight.com/index.php?topic=39004.msg1458478#msg1458478
[21:19] <Laurenceb> dat sarcasm
[21:32] DL1SGP (~felix64@dhcp6.signon3.dk.beevpn.com) left irc: Ping timeout: 255 seconds
[21:34] DL1SGP (~felix64@217.15.160.134) joined #highaltitude.
[21:39] g0hww (~g0hww@46.18.105.34) left irc: Ping timeout: 272 seconds
[21:44] SA6BSS-Mike|2 (~SA6BSS-Mi@81-236-232-232-no30.tbcn.telia.com) left irc: Write error: Connection reset by peer
[21:44] SA6BSS-Mike|2 (~SA6BSS-Mi@81-236-232-232-no30.tbcn.telia.com) joined #highaltitude.
[21:45] <fl_0> re
[21:45] <fl_0> who is the author/owner of the x-f.lv database cockpit telemetry stuff ? :)
[21:46] <mattbrejza> x-f is
[21:47] Action: Upu points at x-f
[21:47] <fl_0> oh i c :)
[21:47] <fl_0> x-f: ping
[21:47] <fl_0> :)
[21:48] <Lunar_Lander> AndyEsser, still on?
[21:48] <Lunar_Lander> there is something coming up in a few
[21:48] <fl_0> I am wondering why our pressure values are not reported to the dashboard
[21:59] gonzo_nb (~gonzo@92.6.244.28) joined #highaltitude.
[22:02] zyp (zyp@zyp.no) left irc: Remote host closed the connection
[22:05] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[22:07] <AndyEsser> Lunar_Lander: I'm around, yes
[22:09] Hix_ (~hix@97e08719.skybroadband.com) left irc: Ping timeout: 246 seconds
[22:14] <Lunar_Lander> thanks
[22:18] <Lunar_Lander> AndyEsser, https://myshare.uni-osnabrueck.de/d/ae5c91202e/
[22:21] zyp (zyp@zyp.no) joined #highaltitude.
[22:36] <AndyEsser> Lunar_Lander: trying to give me neck pain?
[22:36] TIBS01 (~tibs01@5751bf78.skybroadband.com) left irc: Ping timeout: 250 seconds
[22:37] <Lunar_Lander> no, I don't know why these are rotated
[22:37] <Lunar_Lander> please download to rotate
[22:37] <AndyEsser> I will do :)
[22:37] <AndyEsser> thanks very much
[22:37] <AndyEsser> what V is your battery btw?
[22:37] <Lunar_Lander> 12
[22:38] <AndyEsser> ah ok
[22:38] <AndyEsser> Lunar_Lander: what's an FTDI?
[22:41] <Lunar_Lander> USB-to-UART chip
[22:41] <Lunar_Lander> made in scotland
[22:41] <AndyEsser> ah ok
[22:41] <AndyEsser> so just plug direct into USB, rather than a USB-to-Serial adapter or something?
[22:42] <nick_> The USB-to-Serial probably is an FTDI chip if your uC doesn't do that for you.
[22:42] <AndyEsser> gotcha
[22:43] <nick_> Or it's a fake FTDI chip, and if you're unlucky Windows will brick it for you.
[22:44] <AndyEsser> heh
[22:44] <AndyEsser> might have to use one of those on my board for easier input
[22:45] <AndyEsser> was thinking of putting an RJ45 type port on and using oldschool router/switch config cables, but usb might just be easier ;)
[22:45] <AndyEsser> anyway, I'm off to bed
[22:45] <AndyEsser> shall catch you chaps tomorrow
[22:50] <Lunar_Lander> good night andy :)
[22:58] <Lunar_Lander> Upu, are you up for a short question?
[23:03] <fsphil> is it about hobbits?
[23:04] <Lunar_Lander> no, the inductor in the ublox trace on habduino/PITS
[23:06] Hix_ (~hix@97e08719.skybroadband.com) joined #highaltitude.
[23:10] Hix_ (~hix@97e08719.skybroadband.com) left irc: Ping timeout: 240 seconds
[23:25] <Lunar_Lander> good night :)
[23:33] <Laurenceb> this is useful http://www.materflow.com/files/Material%20Data.pdf
[23:33] <Laurenceb> everything you could ever want to know about laser sintered nylon
[23:33] gonzo_nb (~gonzo@92.6.244.28) left irc: Remote host closed the connection
[23:35] MoALTz (~no@78-11-180-214.static.ip.netia.com.pl) left irc: Ping timeout: 250 seconds
[23:37] MoALTz (~no@78-11-180-214.static.ip.netia.com.pl) joined #highaltitude.
[23:40] BrainDamage (~BrainDama@unaffiliated/braindamage) left irc: Read error: Connection reset by peer
[23:42] chris_99 (~chris_99@unaffiliated/chris-99/x-3062929) left irc: Quit: Leaving
[23:42] BrainDamage (~BrainDama@unaffiliated/braindamage) joined #highaltitude.
[23:51] MoALTz (~no@78-11-180-214.static.ip.netia.com.pl) left irc: Read error: Connection reset by peer
[23:55] bertrik (~quassel@rockbox/developer/bertrik) left irc: Remote host closed the connection
[23:55] Lunar_Lander (~kevin@p54889688.dip0.t-ipconnect.de) left irc: Quit: Leaving
[00:00] --- Thu Dec 17 2015