<< 28-02-2014 >>
00:05:16*Demos_ joined #nimrod
00:14:59*darkf joined #nimrod
00:15:02Araqgood night
00:16:13Discolodai like darcs, but there is nothing like github for it
00:17:27*Matthias247 quit (Read error: Connection reset by peer)
00:21:50Demos_darcs hub is OK, as many repos as you like
00:23:49*askatasu1 joined #nimrod
00:25:34filwiti'm taking off, later folks
00:25:38*filwit quit (Quit: Page closed)
00:26:19*Demos joined #nimrod
00:26:32Demoswow ubuntu is totally unstable
00:26:37*zielmicha quit (Quit: Connection closed for inactivity)
00:26:59*Demos_ quit (Ping timeout: 272 seconds)
00:32:06*EXetoC joined #nimrod
00:34:43skrylarDemos: i'm about to nuke my ubuntu and put win7 back on the box
00:35:13*phI||Ip_ quit (Ping timeout: 240 seconds)
00:35:28*phI||Ip joined #nimrod
00:36:14skrylari spent a while toying with music production on linux and i can still conclude linux is not suitable for production work of any kind that doesn't involve being a server :/
00:36:40*Tyler__ joined #nimrod
00:37:10skrylarfor instance, the audio engineers don't understand what "real time" is lol
00:38:04EXetoCas in kernel latency?
00:38:04Demosyeah I am running ubuntu from a btrfs subvolume, but arch is rock solid in that setup
00:38:11skrylarEXetoC: no, in general
00:38:32EXetoCright
00:38:43skrylarEXetoC: you can get a realtime kernel, JACK supports realtime, it will happily do smart things
00:39:01skrylarand then people go do stupid shit like write vModSynth directly to ALSA and don't bother to separate the GUI/DSP code
00:39:04*OrionPK quit (Read error: Connection reset by peer)
00:39:13*CarpNet quit (Ping timeout: 240 seconds)
00:39:14skrylarSo the DSP code is tripping interrupts/potential mallocs here and there
00:39:47skrylarall stuff you know not to do if you spend 3 seconds reading the VST docs for windows, or even 10 seconds reading the LADSPA/DSSI headers for linux
00:40:04*CarpNet joined #nimrod
00:40:07*Tyler_ quit (Ping timeout: 240 seconds)
00:40:49*Varriount quit (Ping timeout: 240 seconds)
00:40:49*icebattle quit (Ping timeout: 240 seconds)
00:40:49*silven quit (Ping timeout: 240 seconds)
00:40:57skrylarSo combine that the best asset for linux audio (JACK) isn't even used by half or more of the audio software available, the audio software is more often than not written incorrectly its ugly to boot
00:41:36*icebattle joined #nimrod
00:41:37*comex quit (Ping timeout: 240 seconds)
00:42:01*brson quit (Ping timeout: 240 seconds)
00:42:13skrylarassuming you persevere all this, tools like Ardour are poorly documented and things like ZynAddSubFX just go "hey good luck at all 200 dials we put in 8 years ago" >_>
00:42:16skrylarimma stop now
00:44:28EXetoC:>
00:45:56*Demos quit (Remote host closed the connection)
00:46:06*vendethiel quit (Ping timeout: 240 seconds)
00:46:16*evil_CarpNet joined #nimrod
00:46:40*vendethiel joined #nimrod
00:47:04*CarpNet quit (Ping timeout: 240 seconds)
00:47:57*brson joined #nimrod
00:48:51EXetoCoh well. some day maybe
00:49:01skrylarsome day after armageddon for sure :)
00:49:03EXetoCI'm just messing about with semi-stable audio tools
00:49:37EXetoCwell renoise is pretty stable, but the waveform editing is limited, which is silly considering how sample-oriented it is
00:49:42skrylarAh renoise is nice
00:49:56skrylarI was going to get renoise and aspect once upon a time, because they're both linux native
00:50:12*Demos joined #nimrod
00:50:23Demosok I am now in a noncrappy IRC client
00:50:31skrylarI don't use renoise as much as I should; its a decent sampler but i don't like how patterns can't have varying length
00:50:38skrylarlike my drumbeat has to be the same length as everything else
00:50:44EXetoCaspect? never heard of it
00:50:59skrylarhttp://www.loomer.co.uk/
00:51:14skrylarits a softsynth
00:52:13EXetoCI'll look into it
00:52:39skrylarEXetoC: i think renoise is nice as far as the tracker style makes it easy to toss in notes via keyboard, but its abbhorrant if you plan to use a midibaord and record live
00:52:53skrylarall though trackers were meant for keyboard entry anyway, so. lol
00:53:53EXetoCAll I know is that there are live-oriented plugins
00:54:03*icebattle quit (*.net *.split)
00:54:05*r0b1 quit (*.net *.split)
00:54:15*icebattle joined #nimrod
00:54:20skrylaryeah i heard good things about the lauflicht-something sequencer for that
00:54:38skrylarseq24 and harmony are also neat for live looping
00:54:59DemosBitPuffin: ping
00:55:00*Zuchto_ quit (Excess Flood)
00:55:05*Zuchto joined #nimrod
00:55:12*cark quit (Write error: Connection reset by peer)
00:56:01BitPuffinDemos: pong
00:56:20Demoslinagl/matrix seems to ICE the compiler, can you try and repro?
00:56:31BitPuffinDemos: hmm
00:57:14*Araq quit (Ping timeout: 240 seconds)
00:57:21BitPuffinDemos: what's the code?
00:57:55*silven joined #nimrod
00:58:09Demosimport linagl/matrix
00:58:10BitPuffinworks for me
00:58:48Demoscompiler verison?
00:58:52BitPuffinlaest git
00:58:54BitPuffinmaster
00:59:13Demoscf81177c407559ce45fa06a20ed91e56ede6aaff
00:59:15Demosfor me
00:59:20DemosI am on devel
01:00:59*carum quit (Remote host closed the connection)
01:03:48BitPuffinDemos: try master
01:04:21Demosmaster breaks other things for me
01:05:37*carum joined #nimrod
01:06:42*cark joined #nimrod
01:08:46*brson quit (Changing host)
01:08:46*brson joined #nimrod
01:11:51DemosI switched to using my own matrix code. I will contribute it to linagl when static[T]s fully work
01:12:56*Araq_bnc joined #nimrod
01:15:17*EXetoC quit (Read error: Connection reset by peer)
01:21:25*Varriount joined #nimrod
01:21:47VarriountHm. Anyone read the discussion on the forum about symbol exportation?
01:23:28*xenagi joined #nimrod
01:24:45*ddl_smurf_ joined #nimrod
01:25:20Demoswe were talking about it a few hours ago, I decided that we dont need it because you can just make a new module and use the export keyword, araq considered it bikeshedding
01:26:02VarriountI have to agree with Araq on that
01:26:17Demosyeah I do too
01:27:00VarriountIf people want wierd access restrictions beyond what's available, they can use macros.
01:27:05*silven_ joined #nimrod
01:27:13*DAddYE_ joined #nimrod
01:27:48*carum quit (Remote host closed the connection)
01:27:59*rixx_weechat joined #nimrod
01:28:38*brson_ joined #nimrod
01:29:18skrylarhow do macros actually help with that
01:29:45*brson quit (Ping timeout: 240 seconds)
01:29:45*ddl_smurf quit (Ping timeout: 240 seconds)
01:29:46*silven quit (Ping timeout: 240 seconds)
01:29:48*rixx quit (Ping timeout: 240 seconds)
01:29:48*DAddYE quit (Ping timeout: 240 seconds)
01:30:57*ddl_smurf_ is now known as ddl_smurf
01:32:31*carum joined #nimrod
01:36:35*carum quit (Remote host closed the connection)
01:36:47*carum joined #nimrod
01:39:07Varriountskrylar: Raise an error if the macro detects that the source file it's being used in isn't right.
01:45:35*carum quit (Remote host closed the connection)
01:59:50*q66 joined #nimrod
02:00:50*carum joined #nimrod
02:03:38*Demos quit (Ping timeout: 240 seconds)
02:05:12*carum quit (Ping timeout: 240 seconds)
02:05:29*q66 quit (Ping timeout: 252 seconds)
02:06:58*DAddYE_ quit (Remote host closed the connection)
02:07:31*DAddYE joined #nimrod
02:11:36*DAddYE quit (Ping timeout: 240 seconds)
02:15:05skrylarVarriount: woot, its nice when you finally see a screen of [OK]'s
02:15:25*zahary joined #nimrod
02:17:38*zahary1 quit (Ping timeout: 240 seconds)
02:18:22BitPuffinI'm not sure what Demos means
02:18:37skrylarBitPuffin: with regards to...?
02:18:48BitPuffinwith contributing his own matrix code
02:19:12skrylarit sounded like he had trouble getting linalg to work and wrote his own matrix code
02:19:39BitPuffinlike I don't know what his goals with it are
02:19:42BitPuffinyeah well
02:19:47BitPuffinIt works over here
02:19:58BitPuffinIs he on Windows?
02:20:32Varriountskrylar: :D
02:21:16VarriountHm. Anyone have ideas on how I should respond to this? -> https://github.com/wbond/package_control_channel/pull/2884
02:21:20*q66 joined #nimrod
02:21:33*q66 quit (Changing host)
02:21:33*q66 joined #nimrod
02:22:31Varriountskrylar: Can I see your pretty unicode tools?
02:22:59skrylarVarriount: https://github.com/Skrylar/skUtf
02:27:26Varriountskrylar: Why the old-style prefixes?
02:30:09skrylarVarriount: the 'sk' thing?
02:31:12Varriountskrylar: No, the "T" thing, like "TCodePoint"
02:32:10*zahary1 joined #nimrod
02:32:11skrylaryou mean we don't do pascal naming anymore? '_'
02:33:16Varriountskrylar: Depends on who you ask.
02:34:35Varriountskrylar: http://forum.nimrod-lang.org/t/191/1
02:34:38xenagiI am not a fan of it
02:37:06*zahary quit (Ping timeout: 240 seconds)
02:38:31skrylari saw TFoo in the reference code and just went with it
02:38:56skrylarugh.. i don't like the case sensitivity
02:39:46Varriountskrylar: Semi-sensitivity actually.
02:40:17skrylari guess i better post on that, because the "style doesn't matter all the code just works" is one of those key things that i found endearing
02:40:18*comex joined #nimrod
02:42:39skrylarVarriount: i don't really see the vendetta against the P-prefix
02:43:09skrylarI get removing the T prefix because 'meh', but how is "BlahPtr" better than "PBlah" or "BlahRef" better than "PBlah" or "RBlah"
02:44:14*onionhammer joined #nimrod
02:44:52Varriountskrylar: I'm not of any opinion.
02:44:54xenagiI find BlahPtr more readable, as though I can actually read it, rather than decode then read it
02:45:15VarriountI don't have enough experience to be fossilized in my ways.
02:45:40onionhammerforgot to run irc in screen though, so when i log off ssh it's going to disconnect
02:45:43onionhammeroops
02:45:49skrylari don't really have a huge opinion on T/P
02:45:59skrylarcase sensitivity is something i would rage over though
02:46:15xenagiyou prefer case insensitivity?
02:46:30skrylarxenagi: it makes life a lot better when dealing with mixed code
02:47:05skrylarConsider that one C library will use_snake_case_like_sdl and then another library DecidesToUseThis and someone else Derp_HERP_MERP()
02:47:15skrylarIn every other language you just have to deal with that shit
02:47:22xenagiyeah
02:47:27skrylarin nimrod its like
02:47:37skrylar"Oh, nice, I can completely *not care* about that person's coding style"
02:47:54xenagiyeah, I never thought about it like that.
02:48:17xenagiPersonally, I wasn't/am not a big fan of case insensitivity because it feels restricting
02:48:31xenagibut in truth, I'd never write two variables that only differ in case
02:48:43skrylari haven't really had very many issues with it, and i don't understand a lot of viewpoints who do
02:49:02skrylarbecause, yeah, having "x" and "X" as two values is already considered bad form
02:56:33VarriountI wonder if case sensitivity is an option you can push/pop with a pragma.
03:11:08skrylari was thinking about suggesting that
03:31:24*DAddYE joined #nimrod
03:34:04skrylarVarriount: i don't see what good it would do though
03:45:27*q66 quit (Quit: Leaving)
04:01:20*zahary joined #nimrod
04:01:20*zahary1 quit (Read error: Connection reset by peer)
04:12:32*brson_ quit (Ping timeout: 252 seconds)
04:21:22*onionhammer quit (Remote host closed the connection)
04:28:53*DAddYE quit (Remote host closed the connection)
04:29:19*DAddYE joined #nimrod
04:33:46*DAddYE_ joined #nimrod
04:34:06*DAddYE quit (Ping timeout: 265 seconds)
04:35:53*comex_ joined #nimrod
04:37:28*comex quit (Ping timeout: 240 seconds)
04:41:46*DAddYE_ quit (Remote host closed the connection)
04:42:15*DAddYE joined #nimrod
04:42:45*silven_ quit (*.net *.split)
04:46:40*DAddYE quit (Ping timeout: 265 seconds)
04:48:50*xenagi quit (Quit: Leaving)
04:52:01*DAddYE joined #nimrod
04:58:43*zahary1 joined #nimrod
04:59:07*Varriount_ joined #nimrod
04:59:36*ddl_smurf_ joined #nimrod
05:01:43*DAddYE_ joined #nimrod
05:02:58*DAddYE quit (Read error: Connection reset by peer)
05:04:15*silven joined #nimrod
05:08:44*cark quit (Write error: Connection reset by peer)
05:09:50*zahary quit (Ping timeout: 286 seconds)
05:09:56*ddl_smurf quit (Ping timeout: 286 seconds)
05:09:56*ddl_smurf_ is now known as ddl_smurf
05:10:01*cark joined #nimrod
05:11:03*Varriount quit (Ping timeout: 245 seconds)
05:27:52*cark quit (Write error: Connection reset by peer)
05:28:05*psquid quit (Ping timeout: 330 seconds)
05:28:07*eigenlicht quit (Ping timeout: 330 seconds)
05:28:12*cark joined #nimrod
05:28:34*psquid joined #nimrod
05:29:07*eigenlicht joined #nimrod
05:48:54*cark quit (Write error: Connection reset by peer)
05:49:01*cark joined #nimrod
06:05:14*iNode001 quit (Ping timeout: 539 seconds)
06:05:14*cark quit (Write error: Connection reset by peer)
06:05:22*iNode001 joined #nimrod
06:05:33*cark joined #nimrod
06:10:48*carum joined #nimrod
06:22:34*Demos joined #nimrod
06:22:34*Demos quit (Client Quit)
06:22:44*Demos joined #nimrod
07:00:43*darkf_ joined #nimrod
07:00:54*darkf quit (Disconnected by services)
07:00:56*darkf_ is now known as darkf
07:07:54*Demos quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
07:16:10*skrylar quit (Quit: leaving)
07:40:54*carum quit (Remote host closed the connection)
08:26:35*Araq_bnc is now known as Araq
09:02:37*DAddYE_ quit (Remote host closed the connection)
09:03:04*DAddYE joined #nimrod
09:07:41*DAddYE quit (Ping timeout: 265 seconds)
09:30:42*EXetoC joined #nimrod
09:50:57*rixx_weechat is now known as rixx
10:00:01*nande quit (Remote host closed the connection)
10:03:36*DAddYE joined #nimrod
10:08:06*DAddYE quit (Ping timeout: 265 seconds)
10:19:29*DAddYE joined #nimrod
10:23:37*DAddYE quit (Ping timeout: 240 seconds)
10:27:10*Mordecai joined #nimrod
10:27:29*Mordecai is now known as Guest60708
10:27:49*psquid quit (Ping timeout: 244 seconds)
10:38:05*DAddYE joined #nimrod
10:42:24*DAddYE quit (Ping timeout: 240 seconds)
11:17:13*r0b1 joined #nimrod
11:38:35*DAddYE joined #nimrod
11:39:51*io2 joined #nimrod
11:43:19*DAddYE quit (Ping timeout: 265 seconds)
12:29:23*faassen joined #nimrod
12:34:40*Guest60708 is now known as psquid
12:39:21*DAddYE joined #nimrod
12:44:13*DAddYE quit (Ping timeout: 265 seconds)
13:04:08*io2 quit ()
13:07:34*io2 joined #nimrod
13:14:36*faassen quit (Quit: Leaving.)
13:18:17*askatasu1 quit (Quit: WeeChat 0.4.2)
13:18:34*askatasuna joined #nimrod
13:25:47*awestroke joined #nimrod
13:33:16*faassen joined #nimrod
13:36:24*awestroke quit (Ping timeout: 244 seconds)
13:36:53*awestroke joined #nimrod
13:37:51*darkf quit (Quit: Leaving)
13:39:53*DAddYE joined #nimrod
13:44:38*DAddYE quit (Ping timeout: 265 seconds)
13:47:29*awestroke quit (Ping timeout: 240 seconds)
13:49:50*q66 joined #nimrod
13:51:44*faassen left #nimrod (#nimrod)
13:58:29*psquid quit (Ping timeout: 240 seconds)
13:59:43*psquid joined #nimrod
14:04:17*OrionPK joined #nimrod
14:16:26*q66 quit (Ping timeout: 252 seconds)
14:20:12OrionPKhttp://dylanfoundry.org/2014/02/28/early-thoughts-on-atom/
14:23:00dom96https://twitter.com/comex/status/439199481515741185
14:23:05dom96https://twitter.com/comex/status/439200487955771392
14:23:13dom96Doesn't sound like it's good performance wise.
14:28:38OrionPKwe'll see
14:28:49OrionPKi havent had time to pay w/ it myself
14:40:32*DAddYE joined #nimrod
14:45:07*DAddYE quit (Ping timeout: 244 seconds)
14:51:23*ddl_smurf quit (Quit: ddl_smurf)
15:11:28*dom96 is now known as nimrod_dom96
15:11:50Tyler__hrrmm...
15:12:02Tyler__So in memory.nim I ahve this line: var memory*: array[0..3999, TWord] (TWord is also exported)
15:12:26Tyler__in another file if I import memory and try to access the memory array I gety a Expression has no type or ambiguous type error
15:18:13*Demos joined #nimrod
15:22:50*nimrod_dom96 is now known as nimrod_dom
15:32:10nimrod_domTyler__: Gist your code.
15:32:48*EXetoC quit (Quit: WeeChat 0.4.3)
15:34:10*Endy joined #nimrod
15:34:23*ddl_smurf joined #nimrod
15:39:22*psquid quit (Quit: work)
15:41:11*DAddYE joined #nimrod
15:45:34*DAddYE quit (Ping timeout: 244 seconds)
15:49:37*[1]Endy joined #nimrod
15:52:25*Endy quit (Ping timeout: 240 seconds)
15:52:25*[1]Endy is now known as Endy
16:00:23*eigenlicht quit (Ping timeout: 252 seconds)
16:01:37*Demos quit (Ping timeout: 240 seconds)
16:21:56*nimrod_dom is now known as dom96
16:22:28dom96http://forum.nimrod-lang.org/t/367#2096
16:22:41dom96Just had the GSoC rejection meeting.
16:22:53dom96It looks like our name wasn't a factor in their decision.
16:23:18dom96bbl
17:18:35shodan45aww, no gsoc?
17:19:00*nande joined #nimrod
17:20:32dom96shodan45: nope
17:25:49reactormonkdom96, you got the reason?
17:26:01dom96Read the forum post I made
17:28:34Discolodamy impression is it was the quality/detail of each idea, and not enough interesting ones
17:28:35reactormonkdom96, how do you find someone to vouch?
17:28:51dom96Discoloda: yep
17:28:59dom96reactormonk: I have no idea
17:29:11reactormonkdom96, soooo why didn't you ask?
17:29:57dom96Because that question didn't come into my mind.
17:30:17dom96I didn't realise I would be able to ask questions.
17:30:18reactormonkduh.
17:30:23dom96Otherwise I would have prepared some...
17:30:39dom96We can ask that next year though.
17:42:44*nande quit (*.net *.split)
17:42:47*ddl_smurf quit (*.net *.split)
17:42:52*cark quit (*.net *.split)
17:43:00*OrionPK quit (*.net *.split)
17:43:00*io2 quit (*.net *.split)
17:43:01*r0b1 quit (*.net *.split)
17:43:04*Raynes quit (*.net *.split)
17:44:12Tyler__dom96: sorry for delay, code is here: https://github.com/tylereaves/nimix
17:44:22Tyler__the error occurs compilign ops.nim on line 7
17:45:57*shodan45 quit (Remote host closed the connection)
17:50:27*Raynes joined #nimrod
17:50:39*shodan45 joined #nimrod
17:50:39*nande joined #nimrod
17:51:31*shodan45 quit (Client Quit)
17:56:25dom96Tyler__: Probably because your array has the same name as your module
17:56:47Tyler__hrrm, interesting
17:56:55Tyler__docs give the impression that the module name doesn't really matter
17:57:05Tyler__I even tried memory.memory[address] and that also fails
17:57:15Tyler__but I suppose I can try renaming the module
17:58:02Tyler__and that seems to have fixed it
17:59:17*carum joined #nimrod
18:03:20Araqhi dom96
18:03:25dom96hey Araq
18:03:43*Varriount|Mobile joined #nimrod
18:04:00Araqdo you have a link to Julia's gsoc application? maybe we can learn from that
18:04:31dom96http://julialang.org/gsoc/2014/
18:04:32*DAddYE joined #nimrod
18:05:58*EXetoC joined #nimrod
18:11:12*EXetoC quit (Ping timeout: 240 seconds)
18:16:28*EXetoC joined #nimrod
18:18:55dom96reactormonk:
18:18:58dom96<MovingBlocks_Cer> yep - hints on how to actually court vouchers is one of my two questions :-)
18:19:07dom96<carols_> well, it's mainly about finding orgs that are working in your space that you can collaborate with
18:19:07dom96<carols_> and then having them attest to how cool you are later
18:20:07*Matthias247 joined #nimrod
18:25:57*Boscop joined #nimrod
18:31:57*cark joined #nimrod
18:38:17*zielmicha joined #nimrod
18:41:13*evil_CarpNet quit (Quit: Leaving)
19:01:17*r0b1 joined #nimrod
19:18:40*q66 joined #nimrod
19:18:40*q66 quit (Changing host)
19:18:40*q66 joined #nimrod
19:18:50*BitPuffin quit (Ping timeout: 264 seconds)
19:19:12*BitPuffin joined #nimrod
19:20:28*awestroke joined #nimrod
19:24:45*q66 quit (Quit: Leaving)
19:25:05*q66 joined #nimrod
19:29:06*carum quit (Remote host closed the connection)
19:29:35NimBotAraq/Nimrod devel b38a1b0 Dominik Picheta [+0 ±1 -0]: Fixes sockets2 on Windows.
19:38:02*carum joined #nimrod
19:48:17*brson joined #nimrod
19:49:46*carum quit (Remote host closed the connection)
19:55:09*carum joined #nimrod
20:11:28*carum quit (Remote host closed the connection)
20:12:27*carum joined #nimrod
20:12:28Araqso ... I decided only babel packages get a prefix for the generated C files
20:16:10dom96I don't think that's a good idea.
20:16:47dom96Is this really so complicated?
20:17:27dom96let prefix = modulePath.replace(PathsThisModuleWasFoundIn)
20:19:29Araqbah
20:22:36*Demos joined #nimrod
20:26:52*nequitans joined #nimrod
20:27:33DemosMatthias247, nimlime question?
20:27:44Matthias247Demos: yep?
20:27:57nequitansAhoy!
20:28:16DemosI am looking at the AutoComplete.py code and I am seeing that you are not using trackDirty. does this cause problems?
20:30:05nequitansi would like to use openmp with nimrod and noticed the || operator
20:30:23Matthias247Demos: i do?
20:30:39Demosoh wait missed it
20:30:41Demossorry
20:30:45Matthias247pargs = pargs + "--trackDirty:" + dirtyFileName + ","
20:31:25Matthias247forgive all coding style, was my first-time python achievement :)
20:31:27nequitansis the idea something like: for i in 0 || 10000: echo i would execute each echo i in parallel?
20:31:42Demosnew file every time it looks like as well
20:31:50nequitansi noticed the compiler flags didn't have openmp, so i'm probably misinterpreting how to do this :)
20:31:52Demoswhich is fine, but kinda annoying
20:32:24Matthias247you mean the creation of a dirty file?
20:32:27Demosyeah
20:32:44Matthias247hmm, we could reduce it to a dirty file which is reused
20:32:57Demosmaybe not worth it though
20:33:10Demossince ideally you would just give the compiler some memory
20:33:21Matthias247but I would hope to get to a "clean" solution anyway - which means sending the content of the dirty file directly to the comiler instead of storing it first
20:33:33Demospipes?
20:33:52Matthias247we already have a pipe - the compilers stdin :)
20:34:08Demosmore pipes?
20:34:33DemosI actually would love all this to be exposed as a API that we could just call
20:34:44Araqnequitans: yes
20:34:48Matthias247no, one is sufficient. The compiler would have to be modified so that it accepts file directly thorugh the input stream
20:35:10Demosbut idetools is accepting commands from there, at least for batch mode
20:35:23Araqhowever|| has been hacked into the language and you *must not* use any code that calls the GC in the for loop body
20:35:46Araqthat includes 'echo i' as that allocates a string for the 'i' currently
20:35:48Matthias247e.g. change to --trackDirtyBytes:NrOfBytes,$file_content_here
20:36:16nequitansAraq, ah, I see
20:36:18Araqthe effect system will ensure that soon, so you'll get a nice compiler error if you screw up
20:36:47Araqbut currently tehe compiler doesn't help you at all when it comes to ||
20:39:14*q66 quit (Ping timeout: 252 seconds)
20:39:46nequitansSo take it if I want to do parallel ops with GC allocated stuff, I should use threads
20:39:53nequitans(for now at least)
20:39:57Matthias247The problems most likely are: Idetools currently reads line-by-line. So we would either have to escape the file data or change that. And that I don't know how difficult it would be to feed a bytestream to the compiler instead of giving it a file name
20:40:20*q66 joined #nimrod
20:49:23reactormonkDemos, organizations?
20:49:32reactormonkehh @ dom96
20:53:38dom96reactormonk: hrm?
20:55:34*betawaffle quit (Ping timeout: 244 seconds)
20:59:34*carum quit (Remote host closed the connection)
21:00:15*betawaffle joined #nimrod
21:00:50*awestroke quit (Ping timeout: 264 seconds)
21:10:44reactormonkdom96, so we need some organization that vouches for nimrod?
21:10:59dom96or a googler
21:20:38reactormonkdom96, so go gay and flirt with some googler :>
21:22:57dom96lolno
21:23:31reactormonkno dedication to the cause?
21:24:51profmakxthere's lots of google people into languages
21:25:47profmakxwhat nimrod needs is more public visibility
21:35:54Matthias247the all-day-on-reddit that rust and haskell have? :)
21:36:35*awestroke joined #nimrod
21:36:37profmakxi am still not really sure whats s ogreat ao
21:36:39profmakxabout rust
21:36:59Araqwe're on reddit too these days
21:37:15*nequitans quit (Remote host closed the connection)
21:37:25Araqit would be nice if you could participate and fight for us
21:38:47*Matthias247_ joined #nimrod
21:40:06Araqapparently we're even "Rust's cousin" now... lol
21:40:45Matthias247profmakx: probably because it's from mozilla ;)
21:41:00Matthias247it sees a lot of work, it compiles directly to native
21:41:09profmakxand it's nothing special
21:41:44Matthias247and borrows many things from functional languages, which attracts the functional guys (where are from my pov always more attracted to new shiny languages)
21:41:49profmakxI guess I'll never come to terms with hypes
21:42:20Matthias247the question is if it can hold and also attract "normal" programmers, or if it will pass by
21:47:57*OrionPK joined #nimrod
21:50:53*Demos_ joined #nimrod
21:53:16*OrionPK quit (Read error: Connection reset by peer)
21:53:17*awestroke quit (Ping timeout: 240 seconds)
21:53:19*brson quit (Ping timeout: 240 seconds)
21:53:19*BitPuffin quit (Ping timeout: 240 seconds)
21:53:20*DAddYE quit (Ping timeout: 240 seconds)
21:53:20*nande quit (Ping timeout: 240 seconds)
21:53:22*Demos quit (Ping timeout: 240 seconds)
21:53:33*DAddYE joined #nimrod
21:53:57*Endy quit (Ping timeout: 244 seconds)
21:54:06*brson joined #nimrod
21:54:23*Matthias247_ quit (Ping timeout: 245 seconds)
21:54:50*BitPuffin joined #nimrod
21:55:06*nande joined #nimrod
21:55:14*r0b1 quit (Ping timeout: 240 seconds)
21:55:26*awestroke joined #nimrod
21:55:56*Matthias247_ joined #nimrod
21:56:29*r0b1 joined #nimrod
22:03:54Demos_hm I have spent 45mins trying to figure out how the hell to find the end of the buffer VS gives me for completion. Does idetools even need anything after the point where the completion is?
22:04:13Araqno
22:04:48Matthias247_Demos_: do you make a language pack for Visual studio with autocomplete?
22:05:03Demos_I am just getting autocomplete working now
22:05:14Matthias247_still cool :)
22:05:30Matthias247_but don't try it with CAAS yet, it's buggy
22:08:32*BitPuffin quit (*.net *.split)
22:08:33*DAddYE quit (*.net *.split)
22:08:38*Matthias247 quit (*.net *.split)
22:08:55Demos_well I need CAAS
22:09:01Demos_oh you mean batch mode or whatever
22:09:06Demos_yeah I am just using idetools
22:09:14*DAddYE joined #nimrod
22:11:14*io2 joined #nimrod
22:12:32*BitPuffin joined #nimrod
22:14:08Matthias247_Demos_: idetools worked for me. But not the --server version where you get a persistent compiler service
22:28:52*awestroke quit (Remote host closed the connection)
22:34:14*Puffin joined #nimrod
22:46:14*BitPuffin quit (Ping timeout: 240 seconds)
22:47:23*q66 quit (Ping timeout: 252 seconds)
22:57:50*nequitans joined #nimrod
22:58:28Araqguys, vote for me: http://www.reddit.com/r/programming/comments/1z6j37/on_garbage_collection_in_games/cfrb3jy
23:01:32nequitansHello all!
23:02:20dom96hello nequitans!
23:03:46nequitansI am now running a parallel operation using threads where I have preallocated a number of large data sets in a {.global.} array. I perform operations independently on each of them in 10 threads. Theoretically, there should be no data copied to the garbage collected heaps, but i'm noticing that the parallel implementation is far slower than the serial implementation. Any ideas as to why this could be?
23:04:12*q66 joined #nimrod
23:05:02*q66 quit (Read error: Connection reset by peer)
23:05:23*psquid joined #nimrod
23:05:41Varriount|Mobilenequitans: We would have to see your code
23:10:14nequitansVarriountIMobile, thanks. I tried to strip it down and created a gist: https://gist.github.com/geetduggal/9281916
23:10:39nequitansIt depends on a simple matrix type i made that is just essentially a seq[T] with nrows and ncols. There's also this inplace shuffling algo that i didn't include
23:19:29Araqnequitans: how does diagShuffleInplace work?
23:19:33Varriount|Mobilenequitans: Perhaps the profiler could help as well?
23:19:50nequitans(Araq, upvoted.)
23:21:21nequitansAraq, updated the gist to show that code
23:22:57nequitans(by the way, it's been a number of months since I've played with Rust, but my experience with Nimrod has been a lot more pleasant. clearly, rust offers a lot of memory safety, but there are a lot of style and design decisions I prefer in nimrod)
23:24:10Varriount|MobileI upvoted too. Go Araq!
23:26:57fowlnequitans, try using as many threads as your processor allows
23:28:35AraqVarriount|Mobile: what about ... you know?
23:28:39NimBotnimrod-code/Aporia master 39ae389 Dominik Picheta [+0 ±2 -0]: Fixed Save As bug.
23:29:23Varriount|MobileAraq: Hm?
23:29:40Araqthe tester needs to test babel packages, Varriount
23:29:58nequitansfowl, running now -- all processors being used near 100%
23:30:03Varriount|MobileAraq: I have the code nearly integrated
23:30:07nequitans(but still slow)
23:30:30Varriount|MobileI *have* been out sick all week with a throat infection, FYI
23:31:11AraqVarriount|Mobile: get well soon!
23:31:30fowlnequitans, i mean the native amount of threads, going abover that is virtual threads
23:32:08Araqthe optimal number of threads seems to be 2* numberOfCores for CPU bound tasks, don't ask me why
23:32:34nequitansyea, i have 16 cores 32 hyperthreaded, so i ran with 32
23:34:44Varriount|Mobile16 Cores? What processor?
23:34:44dom96http://reza.jelveh.me/2014/02/28/a-critical-look-at-atom.html
23:37:18nequitansVarriountIMobile, 2x8 core at 2.6GHz/core Intel Xeon
23:38:18nequitansThanks for your help guys -- I gotta jet out, but will try to install a mobile IRC client on my phone to keep up
23:41:33Araqnequitans: you might have some false sharing between cpus
23:41:50nequitansinteresting
23:41:53Araqwhich kills performance but still should be faster than the single threaded version, so I dunno
23:42:51nequitansyea, i think i'm doing something stupid somewhere -- or it's something a lot more subtle
23:43:40Araq let j = random(i+1) # could be your bottleneck too
23:43:59Araqoh and you use -d:release, right?
23:44:46Araqline 31
23:44:50Araqid+i # wtf
23:44:59*OrionPK joined #nimrod
23:46:51Araqthat has to be id*numMatPerThread+i - 1 or something
23:48:42*tinAndi_ joined #nimrod
23:49:27Araqnequitans: you don't split up the work properly, so the threads step on each other's toes
23:55:08*tinAndi_ is now known as tinAndi
23:55:16*Varriount|Mobile quit (Remote host closed the connection)
23:55:28*Varriount|Mobile joined #nimrod