<< 22-10-2013 >>
00:11:41*zanget joined #nimrod
00:15:23Demoshow does one convert a cstring into a normal string>
00:16:04fowl$somecstring
00:23:02*wlhlm quit (Ping timeout: 240 seconds)
00:24:06Demosthanks, also how the hell do refs work with C interop, will returing a ref PFoo from a proc that is exportc'd return a pointer (that could get GC'd out from under you)?
00:25:36fowli am not sure
00:27:52Demoshm I will just GD_addref if, cast it to a ptr, then have a cstyle close function, and I will write a class to let .net's GC deal with it
00:38:38*fowl quit (Ping timeout: 264 seconds)
00:41:36*gdos quit (Read error: Connection reset by peer)
00:58:06Demosbeing able to wire up the two GCs to talk to each other would be pretty great D:
01:19:39*Associat0r quit (Read error: Connection reset by peer)
01:20:48*Associat0r joined #nimrod
01:20:58*Associat0r quit (Changing host)
01:20:58*Associat0r joined #nimrod
01:24:48*fowl joined #nimrod
01:28:37*DAddYE quit (Remote host closed the connection)
01:30:12*Associat0r quit (Quit: Associat0r)
02:01:22*fowl quit (Quit: Leaving)
02:06:23*gdos joined #nimrod
02:18:51*dorei quit ()
02:22:26*xenagi quit (Ping timeout: 264 seconds)
02:29:45*DAddYE joined #nimrod
02:34:16*DAddYE quit (Ping timeout: 260 seconds)
02:40:22*brson quit (Read error: Operation timed out)
02:40:30*Amrykid quit (Excess Flood)
02:40:46*Amrykid joined #nimrod
02:59:19*Amrykid quit (Changing host)
02:59:19*Amrykid joined #nimrod
03:16:14*gdos quit (Remote host closed the connection)
03:30:30*DAddYE joined #nimrod
03:35:02*DAddYE quit (Ping timeout: 264 seconds)
03:48:17*OrionPK quit (Quit: Leaving)
04:09:57*Endeg quit (Read error: Connection reset by peer)
04:31:21*DAddYE joined #nimrod
04:36:07*DAddYE quit (Ping timeout: 260 seconds)
05:34:01*DAddYE joined #nimrod
05:39:31*DAddYE quit (Ping timeout: 260 seconds)
05:47:34*Demos quit (Quit: Leaving)
06:18:19*DAddYE joined #nimrod
06:18:30*Associat0r joined #nimrod
06:18:30*Associat0r quit (Changing host)
06:18:30*Associat0r joined #nimrod
06:39:17*q66 joined #nimrod
07:18:28*Araq_ joined #nimrod
07:27:11*dyu_ joined #nimrod
07:50:01*q66 quit (Ping timeout: 268 seconds)
08:02:37*wlhlm joined #nimrod
08:28:24*q66 joined #nimrod
08:45:01*Araq_ quit (Read error: Connection timed out)
08:45:47*Araq_ joined #nimrod
09:59:42Araq_hi zanget, welcome
10:10:12*Araq_ quit (Quit: ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258])
11:38:17*faassen joined #nimrod
11:45:40*Araq_ joined #nimrod
12:25:10*BitPuffin joined #nimrod
12:56:39*Ricky_Ricardo joined #nimrod
13:03:00*io2 joined #nimrod
13:19:12*faassen quit (Quit: Leaving.)
13:42:14*Araq_ quit (Quit: ChatZilla 0.9.90.1 [Firefox 24.0/20130910160258])
13:57:11*DAddYE quit (Remote host closed the connection)
13:57:38*DAddYE joined #nimrod
14:02:05*DAddYE quit (Ping timeout: 245 seconds)
14:04:13*Ricky_Ricardo quit (Quit: Ricky_Ricardo)
14:24:36NimBotAraq/Nimrod master 23044fc Mark Flamer [+0 ±1 -0]: remove extra white space
14:24:36NimBotAraq/Nimrod master a1c4bc2 Mark Flamer [+0 ±1 -0]: more whitespace removal
14:24:36NimBotAraq/Nimrod master bb6a172 Simon Hafner [+0 ±1 -0]: Merge pull request #630 from mflamer/master... 2 more lines
15:05:51*Hannibal_Smith joined #nimrod
15:17:59*fowl joined #nimrod
15:40:47*q66_ joined #nimrod
15:41:14*q66 quit (Disconnected by services)
15:41:18*q66_ is now known as q66
15:57:22*dyu_ quit (Quit: Leaving)
16:17:28*q66 quit (Ping timeout: 260 seconds)
16:44:46*Hannibal_Smith quit (Quit: Sto andando via)
16:48:58*q66 joined #nimrod
17:00:03*zanoni quit (Remote host closed the connection)
17:01:14*Endy joined #nimrod
17:02:28*brson joined #nimrod
17:11:08*DAddYE joined #nimrod
17:12:17*DAddYE quit (Remote host closed the connection)
17:12:31*DAddYE joined #nimrod
17:12:37*DAddYE quit (Remote host closed the connection)
17:19:18*zanoni joined #nimrod
17:23:05*DAddYE joined #nimrod
17:41:27*shodan45 quit (Quit: Konversation terminated!)
17:48:00*OrionPK joined #nimrod
17:48:14OrionPKhola.. think openSSL may be borked
17:48:51OrionPKsockets.nim(318, 15) Error: type mismatch: got (PSSL, TSocketHandle) but expected one of: openssl.SSL_set_fd(ssl: PSSL, fd: cint): cint
17:49:13OrionPKsomeone changed all the cints to TSocketHandle but didnt update openssl.nim?
17:49:13Araqthat's a regression due to Varriount's changes I think
17:49:21Araqexactly
17:49:22OrionPKthats what I was thinking
17:49:39OrionPKthis also breaks babel
17:50:01Araqwell I said it's a breaking change
17:50:37Araqso ... please fix it and make a pull request
17:51:02OrionPKim at work atm, I'll leave it to Varriount to fix
17:53:33Araqis this on windowsß
17:53:34Araq?
17:54:12dom96It's probably both since TSocketHandle is distinct.
17:54:56dom96Looks like NimBot should announce when a test fails because there is an SSL test and I bet its failed.
17:55:25Araqyou know what also would help? a "diff" feature for nimbuild ... :P
17:56:55*zanoni quit (Read error: Connection reset by peer)
18:01:22*shodan45 joined #nimrod
18:05:46*zanoni joined #nimrod
18:11:34*zanoni quit (Remote host closed the connection)
18:13:32OrionPKAraq yes I'm on 8/8.1
18:14:33*zanoni joined #nimrod
18:23:47*MFlamer joined #nimrod
18:27:21*fowl quit (Ping timeout: 272 seconds)
18:41:10*q66 quit (Quit: Leaving)
18:41:21*q66 joined #nimrod
18:42:32*faassen joined #nimrod
18:43:33*fowl joined #nimrod
19:25:50*MFlamer quit (Ping timeout: 245 seconds)
19:32:26Araqdamn, I missed MFlamer
19:34:00fowlOrionPK, everyone that ive talked to about windows 8 hates it with passion
19:39:07OrionPKthat sucks for them I guess
19:39:15*Hannibal_Smith joined #nimrod
19:39:22OrionPKeveryone you've talked to about windows 8 has also probably never used it longer than 5 minutes :p
19:39:50Hannibal_SmithI use it as "main OS" and don't like it
19:40:24fowlOrionPK, so it takes 5 minutes to appreciate it?
19:41:10OrionPKfowl no, opinions are formed without even trying it
19:41:41OrionPKnot sure what people don't like about it that is so different from windows 7
19:41:50Hannibal_SmithWindows 8 apps are craps, so is a good OS if you don't use the "Metro side"
19:41:52OrionPKactually, I take it back, I know exactly why
19:41:58OrionPKbut its a dumb reason
19:42:06Araqthe UI has clearly been designed for touchscreens but it's used mostly for PCs
19:42:07OrionPKbecause you can spend .5 seconds a day in metro land
19:42:15Araqthat's why
19:42:52OrionPKthe start menu is designed for touchscreen
19:42:59Araqyeah
19:43:00OrionPKbut if you spend a lot of time in the start menu
19:43:05OrionPKthen... you have other problems
19:43:06OrionPK:)
19:43:19OrionPKI hit 'start' type what I want & hit enter, total time spent = .5 seconds
19:43:19Araqsure but it's enough to not want to look at it
19:43:34OrionPKto each his own, I like the look of it
19:43:45OrionPKdont just stop to look at it though, I've got work to do
19:44:01OrionPKalso it is a pretty good touch interface if you've ever used it on a tablet
19:44:12OrionPKsnapping and pinning apps side by side is definitely a leg up on the competition
19:44:31Araqquite possible but it's primarily an OS for PCs
19:44:38OrionPKthats true
19:44:40Hannibal_SmithI can't find how...on a desktop
19:44:49OrionPKyou cant find how to pin apps on a desktop?
19:45:06Hannibal_SmithHow keep two app open is something great
19:45:09OrionPKdrag the window to the side of the screen
19:45:13OrionPKoh
19:45:21OrionPKmultitasking?
19:45:32OrionPKthought the benefits would be pretty obvious
19:45:33fowlOrionPK, i talked to people, IRL, who bought new computers (regular consumers, not developers)
19:45:43OrionPKyeah regular people are dumb, I agree
19:45:46Araqand it would be nice if Microsoft would simply acknowledge it finally
19:45:49fowlthey all wanted to get windows 7 back
19:46:04OrionPKI cant even look at 7 anymore, looks ugly
19:46:06OrionPKi liek the flat look
19:46:24Hannibal_SmithOrionPK, today all mobile OS can do multitasking
19:46:38OrionPKnot really
19:46:48OrionPKthey can switch from fullscreen one app to another app
19:47:10OrionPKwerent you just arguing that full screen apps are bad?
19:47:35Hannibal_SmithOrionPK, on a desktop, yes
19:48:25Hannibal_SmithI don't use a tablet for something "intensive", so for me don't really matters
19:48:33OrionPKimo, I'd rather have the ability to unpin apps into traditional windows that the current status quo for 8
19:48:49OrionPKbut it's not a big deal, certainly not big enough for all these people to scream about it
19:49:04OrionPKif you dont like how the store apps behave, dont use them
19:49:11OrionPKthe rest of the OS is much improved
19:49:28OrionPKin terms of stability and speed, boot up time, etc
19:49:33Hannibal_SmithThe problem with Windows 8 apps is different IMHO...they are craps
19:49:55OrionPKwindows 8 apps are the only ones that default to fullscreen
19:49:59Hannibal_SmithNo feature rich, graphically horrible
19:50:26Hannibal_SmithI can't see why someone would use a Windows 8 app instead of a "classic desktop" app
19:50:51OrionPKbecause they're on a tablet
19:51:10Hannibal_SmithOrionPK, so why not get an iPad
19:51:18guaquahow should i approach multi-dimensional (in this case 2D) arrays? seq[seq[float]] seems to work
19:51:19Hannibal_SmithThat have better mobile apps?
19:51:22OrionPKbecause ipad sucks at multitasking lol
19:51:23OrionPKI sometimes use facebook or the mail app on my windows 8 desktop
19:51:32OrionPKsnapped to the left or right side of the screen
19:51:36OrionPK400 pixels wide or something
19:51:56guaquai don't know the exact dimensions until runtime, so the builtin arrays don't cut it
19:52:41Araqguaqua: seq[seq[float]] works but seq[float] where you compute the index by hand via y*width+x is faster
19:53:03guaquaokay
19:53:19guaquai'll try that and see
19:54:15Araqwith 0.9.3 you can overload proc `[]`*[T](s: seq[T], x, y: int): var T
19:54:33Araqand use it as a[x,y]
19:55:04guaquait's already out?
19:55:26Araqno 0.9.3 is always latest github
19:55:31guaquaokay
19:55:41guaquai might try that instead
19:55:47Araqwell odd numbers are github versions, even numbers stable releases
19:55:57guaquaeasier to reason about in the code if it's not cluttered with array indexing code
19:55:58Hannibal_SmithLittle OT: http://cmr.github.io/blog/2013/10/21/on-stack-safety/ little reference to Nimrod
19:56:08AraqHannibal_Smith: yeah I noticed
19:56:22OrionPKanybody who comes here should use the github version or they'll be confounded features people are talking about that aren't in the last release
19:56:58Hannibal_SmithIMHO their concurrency plans (Rust) are going to "ruin the language"
19:57:24Araqinteresting, what makes you say that, Hannibal_Smith ?
19:57:57Hannibal_SmithAraq, supporto for light task impose to use all hacks for the function stack
19:57:58dom96OrionPK: IMO Microsoft should stop trying to combine both Windows for tablets and Windows for PC.
19:58:35OrionPKdom96 too late :p
19:58:40dom96That said I haven't tested Windows 8 yet.
19:58:47Hannibal_SmithAnd Araq I can't find other user case for having a great number of task that is not a socket server
19:59:05dom96OrionPK: It's never too late. Btw did you get Jester to work?
19:59:09OrionPKdom96 but I think that it was their best bet to get into the tablet space, they wanted to have the ecosystem
19:59:09Hannibal_SmithA normal OS can today handle a good number of threads
19:59:31Araqyeah unfortunately "socket servers" are quite common though
19:59:32OrionPKdom96 no I didnt, i was going to try it out on my work machine today but ran into the openSSL issue I mentioned before
19:59:48AraqI prefer the asyncio + processes way though
19:59:56Hannibal_SmithYes, I feel the same
20:00:15Araqif only nimrod's asyncio would be complete ...
20:00:35guaquaerlang has some very interesting solutions, especially when it comes to reasoning how the code works, where it blocks etc
20:01:03Hannibal_SmithIMHO Haskell is a level above all others with STM
20:01:07AraqI never understood how you can reason in a dynamic language
20:01:08dom96Araq: Hannibal_Smith: Somebody should do a benchmark of a socket server with 1 lightweight thread per connection vs. <number of cores> threads to handle the load. Perhaps somebody already has done so.
20:01:10guaquaother than that, lightweight process implementations don't seem that interesting
20:01:55Hannibal_SmithAh...Haskell also support light task
20:02:17guaquaAraq: erlang process scheduling is deterministic, well-defined. that's the advantage. there's no question about where the execution of a process is when there's a context switch. don't know how that's related to it having dynamic typing?
20:02:18Hannibal_SmithBut for Haskell...they seems like the only good way
20:02:45Hannibal_Smithdom96, yes, it would be interesting
20:02:47dom96OrionPK: Yeah, but they kinda failed haven't they? Most Windows applications do not work on tablets because they are not compatible with ARM CPUs, the others do not integrate well with the touch interface anyway.
20:03:00guaquadom96: it all boils down to what kind of things you do for each client
20:03:38Araqguaqua: reasoning in dynamic wonderland is always like "ok, let's assume nobody overwrote this built-in operation" ... and gets downhill from that point
20:04:14OrionPKdom96 they failed to make a big entrance into the tablet space, but they couldn't have known they would from the start
20:04:15Araqit's always based on wishful thinking
20:04:33guaquaAraq: well, you don't do that in every dynamic language. (and its despised in pretty much all the reasonable ones...)
20:04:35OrionPKdom96 I wouldnt say most windows apps dont work on tablets either
20:05:03OrionPKdom96 I have three windows 8 tablets (from BUILDs), all 3 run windows 8 apps and windows applications generally, because they're all x86
20:05:17OrionPKdom96 even have nimrod compiler on them all :P
20:05:26Araqguaqua: no it's not despised, it's how the languages work ("override __hash__ and __equals__ and the dictionary works with your type")
20:05:32profmakxat
20:05:55*Hannibal_Smith quit (Quit: Sto andando via)
20:06:18*Hannibal_Smith joined #nimrod
20:06:28OrionPKdom96 the ARM tablets were the major mistake, since x86 chips can already be low power enough to make the whole fiasco a complete waste of time
20:06:51dom96OrionPK: Well since I am just speculating, I will trust your judgement since you have the hardware and have tested Windows 8 whereas I haven't :)
20:06:53OrionPKdom96 they shouldnt have even gone down the road of windows RT
20:06:55Hannibal_SmithI did read at least two people that work on Goole on a large Python codebase
20:07:11OrionPKdom96 you're thinking of windows RT probably,
20:07:13guaquaAraq: well, you are talking about some specific subset of programming languages that i wouldn't call "dynamic". but let's leave this at this
20:07:29Hannibal_SmithThey all says that large Python codebase is horrible in maintenance
20:07:35Hannibal_SmithRefactoring is hard
20:07:49Hannibal_SmithAnd that they are evaluating to migrate the code to Go
20:07:49OrionPKdom96 that wont run desktop applications because it's locked down to a microsoft certificate, whereas the tablets I have (surface/acer and samsung) are all x86 running the full version of windows 8
20:08:31Hannibal_Smith*Google
20:08:51dom96OrionPK: Yeah, I was referencing Windows RT. I liked the idea of the surface but felt that the x86 version was way too expensive and that I may as well get a laptop.
20:09:18OrionPKyeah, surface is essentially a laptop
20:09:25OrionPKwith a floppy keyboard :P
20:09:45Araqguaqua: fair enough I guess
20:10:09*MFlamer joined #nimrod
20:10:21Araqyay, wb MFlamer
20:10:22dom96OrionPK: Yeah, it looks like it would be annoying to type on without a desk.
20:11:06OrionPKyeah, it is. if it's in your lap you definitely would use it in tablet mode rather than 'laptop' mode
20:13:34*Dirkson joined #nimrod
20:13:45Araqhi Dirkson welcome
20:13:58DirksonAraq: Hi there! Thanks : )
20:14:05dom96hello Dirkson!
20:14:32DirksonI'm a C coder looking for a language for a hobby project, and I happened across nimrod on a site doing benchmarks. It looks like kind of a neat idea!
20:14:59*fowl quit (Read error: Connection reset by peer)
20:15:09DirksonI'm a little leary of the garbage collector - I like the assurances of speed, but I'm really used to thinking of them as slow, lumbering things. Are there any articles about Nimrod's GC?
20:15:14MFlamerThanks!
20:15:51Araqthere is only http://nimrod-code.org/gc.html
20:16:12DirksonAraq: Still more than I found ^^
20:16:45Araqyou can avoid the GC though and yes more easily than in D. IMO anyway
20:16:55*Endy quit (Ping timeout: 272 seconds)
20:17:05DirksonAraq: Ahhh, that's another bit of information I was looking for ^^
20:18:13AraqDirkson: you should use the latest github version and might play with --os:standalone. no stdlib then, OSes can be written in this mode, targeting 16 bit cpus is reasonable, etc.
20:18:32DirksonInteresting!
20:19:05DirksonBut my hobby project is a little higher - I'd like to target windows/linux from linux, while using opengl.
20:20:04DirksonThe windows build support should be almost trivial with nimrod - Just swap out "gcc" for mingw's gcc, like I do with my current C project : )
20:21:03Araqopengl + nimrod even got special love in the code generator, so that dead code elimination works nicely with OpenGL
20:21:16DirksonCool.
20:21:17MFlamerI build Nimrod exe's on windows daily. It's trival with mingw, which is included
20:21:26*fowl joined #nimrod
20:21:52DirksonMFlamer: I actually build /from/ linux, though, which is a hair trickier. Normally I have to tell the make process to use different binaries than the standard gcc, ld, etc.
20:22:09MFlamerOK, I see
20:22:57DirksonNormally I get my opengl context from glfw, which I don't think nimrod has a library. Couple questions spring to mind - Does nimrod have any way of interfacing with C libraries? How do nimrod opengl applications normally get their context?
20:23:30dom96Dirkson: https://github.com/EXetoC/nim-glfw3/ :)
20:23:38*peter9477 left #nimrod (#nimrod)
20:23:39Dirksondom96: <3 ^^
20:24:25DirksonWell, shucks. This sounds almost too good. Does Nimrod have any disadvantages?
20:24:45Araqyes the compiler is not nearly as stable as I would like
20:25:02Araqon the other hand, stay away from too much meta voodoo and it's perfectly fine
20:25:10DirksonAraq: Do you mean it's likely to change under me, or it's likely to crash under me?
20:25:28Araqa bit of both :P
20:25:31Dirkson*laughs*
20:27:08MFlamerNot much though. Most of my issues with it have to do with me really pushing the boundaries in weird ways :-)
20:27:30dom96The first code I wrote in Nimrod 4 years ago still compiles today :P
20:27:41Araqdom96: wow. awesome
20:28:02DirksonI doubt I'll be pushing much. Mostly I'm tired of the programming inefficiency of C (~30,000 lines will do that to a guy) and would like a little side project in something a little easier to code with : )
20:28:03MFlamerI think you could tackle some pretty complex projects without much trouble
20:28:15dom96Araq: Yep. You're doing a great job at keeping things compatible even though this is still pre-1.0!
20:28:52DirksonMFlamer: I take it you're the fellow behind the language?
20:29:06MFlamerNo Araq
20:29:34MFlamerI'm just like you, but maybe a few months ahead
20:29:37DirksonAhhh : )
20:30:20MFlamerBut, Im seriously addicted already.
20:31:57DirksonWhat's aporia like? No chance it uses vim keybindings? ^^
20:32:21dom96Dirkson: Nope, sorry. That will come in a later version though.
20:33:11Dirksondom96: Hey, neat ^^ I'm pretty addicted to vim at this point. I don't use huge swathes of what it does... But the ability to, say, instantly replace X with Y on the next 300 lines is remarkably useful.
20:34:36dom96Dirkson: I never got into vim. But this should work well for you: https://github.com/zah/nimrod.vim
20:35:01DirksonYes. Yes it should. Thank you : )
20:35:01MFlamerSublime Text is my env of choice for Nimrod at this point
20:36:28DirksonMFlamer: That does look nice.
20:37:43MFlamerThere are some syntax coloring setups for nimrod kicking around. I have one to share and others do also.
20:38:36DirksonClosed source makes me a little leery, for a utility. I wonder if I could talk him into using my license ^^
20:39:57*BitPuffin quit (Read error: Connection reset by peer)
20:40:22*BitPuffin joined #nimrod
20:40:33DirksonI am not getting the concept of comments-as-syntax. A comment can either be associated with a particular line of code, or not. I get the difference in dealing with it on the backend, but what's the difference for me the developer?
20:44:10AraqDirkson: mostly some early wtf moments until it's a non-issue ;-)
20:44:20Dirkson*laughs*
20:44:40Araqobviously it's done for "nimrod doc"
20:45:22Araqbut I may further relax the rules as the planned AST rendering features don't work out
20:45:34DirksonAST rendering?
20:45:58*filwit joined #nimrod
20:46:20filwitello, #nimrod
20:46:27DirksonHi filwit : )
20:46:41filwithi, Dirkson. haven't seen you around before :)
20:46:47AraqDirkson: sorry, too complex to explain
20:46:48dom96hey filwit, how're ya?
20:46:58filwitdom96: good good
20:47:16filwitdom96: been working on a little diagram for my hymn design
20:47:27dom96filwit: cool
20:47:41filwitdom96: doesn't have the mode stuff yet (which is the really cool part), but i'm liking it so far
20:47:48filwitone sec, i'll upload and show you
20:49:13DirksonPardon the pause. Paramedics moving slowly at the house next door. That's not normally a good thing.
20:49:52Dirksonfilwit: Aye. Just popped on. Considering using Nimrod for a little hobby project I have. So far it looks really neat : )
20:50:48filwitDirkson: cool. glad to see another potential Nimrod coder :)
20:51:03filwitDirkson: what's your project about?
20:51:17Dirksonfilwit: Familiar with falling sand games?
20:51:38dom96ooh I love those!
20:51:40filwitnot really
20:51:51filwitbut i like games in general :)
20:52:08fowlDirkson, use sdl2 instead of glfw
20:52:08dom96I spent more time than I'd like to admit on a couple of those falling sand games...
20:52:34fowlthen u can skip all the opengl bs
20:52:51Dirksonfilwit http://powdertoy.co.uk/#introCarousel - I THINK this is a good example of one.
20:52:52filwitfowl: sld2 up to date?
20:53:06filwitfowl: i'll probably be using it if it is.
20:53:10fowlafaik
20:53:16filwitk
20:53:34DirksonI basically want to make one of these, but base it around gravity and life - Set up a little solar system, add materials to planets, watch life colonize from planet to planet. A little 2d solar system god game, but based on pixel physics, like sand games are.
20:53:40filwitDirkson: looks interesting
20:54:06dom96I used to play this one: http://dan-ball.jp/en/javagame/dust/
20:54:23Dirksonfowl: I've considered that. I figured opengl might be speedier - Set up a texture, blit the texture to the screen. It's been some years since I fuddled with sdl, but I work with opengl nearly every day.
20:54:40Dirksondom96: I believe that's the original : )
20:54:49dom96Dirkson: cool
20:54:53fowlDirkson, sdl2 is a vast improvement
20:55:57Dirksonfowl: Remember that I'd basically be doing single-pixel draws most of the time. Does SDL2 handle that efficiently? I seem to remember it become somewhat slow in SDL1
20:56:07*BitPuffin quit (Quit: WeeChat 0.4.2)
20:56:37filwitdom96: i added another slide to the bottom: http://reign-studios.net/philipwitte/hymn/
20:56:39DirksonOh, good, the neighbor appears alive. Paramedics moving slowly because it's not critical, rather than because of forgone conclusion.
20:56:47AraqDirkson: I can't imagine that to be of any concern on any hardware that's less than 10 years old
20:57:03DirksonAraq: Fair enough : )
20:57:34Dirksonfilwit: You are currently designing this?
20:57:42filwitDirkson: yes :)
20:57:44fowlDirkson, yes because its closer to opengl/directx now, though software renderer is still available
20:57:55Dirksonfilwit: You're unlikely to get it done by February, right? :D
20:58:05filwitDirkson: haha, i wish
20:58:32Dirksonfilwit: I start teaching a kid's programming course in February. Currently I'm still figuring I'll do it in python, but I do need to find some sort of simplified IDE to make it easier for them.
20:58:51dom96filwit: That is sexy :D
20:58:55filwitDirkson: still in design phase though, my brother an I have been working on a C# SDK for awhile (it's mostly his project), here: https://github.com/reignstudios/ReignSDK
20:59:16filwitDirkson: some of the designs will be from that code-base, though obviously changed to fit Nimrod
20:59:25filwitdom96: glad you like it :D
20:59:30Dirksonfowl: Araq: Cool. Based on your feedback, I think I'll just use sdl2 then : )
21:00:15Dirksonfilwit: Not a big fan of C#. The whole mono thing on linux is.... painful. The only associations I have with C# are pain, basically :D
21:00:45filwitdom96: it illustrates how the code will be wired up a little more (it's basically Unity3D, but Nimrod, and more powerful). there are some really cool ideas i have about a built-in state machine using Nimrod's macros for nice syntax (like 'part'), which is something Unity3D is missing by default
21:01:20dom96filwit: will this be open source?
21:01:21filwitDirkson: yeah, C# can be a pain sometimes, but I have to agree with my brother somewhat.. it really is the most practical cross-platform language right now
21:01:45filwitDirkson: as far as a balance between productivity tools and performance... well, i guess there's Java too
21:01:54filwitdom96: of course
21:02:24dom96filwit: Awesome. I will help you as much as I can :)
21:02:34filwitdom96: hoping you'll help me with the project eventually, once the base code gets built up a bit
21:02:42filwitdom96: :D
21:02:45dom96^^ :)
21:02:52*Dirkson guestures at his cross-platform C89 game, which feels appropriate to mention here : ) https://www.scrumbleship.com/
21:03:23fowlneat
21:03:29fowlport that plz
21:03:30filwitnice, Dirkson
21:03:39AraqDirkson: omg you invented minecraft!
21:03:41Dirksonfowl: Port to what? : )
21:03:44filwitand good to see another indi game-dev interested in Nimrod
21:03:50*BitPuffin joined #nimrod
21:04:02DirksonAraq: *laughs* It's not very minecraft. Visually similar, but look at the voxel damage stuff - We're rendering in a completely different way.
21:04:05fowlDirkson, just rewrite in nimrod
21:04:18fowl:p
21:04:25filwitDirkson, and yeah, that's a pretty cool little game
21:04:38DirksonThanks : )
21:05:21fowlhalf of nimrods users must be indie game devs now
21:05:32DirksonMostly pointing it out to prove it really is possible to realistically do cross platform coding without C# and friends... maybe not exactly /sane/, but quite possible!
21:05:33dom96Dirkson: Very nice. I agree with Araq, it's like a space Minecraft :P
21:06:32filwitbtw, Araq: is there any performance overhead of using 'var T' params when the param doesn't actually get modified?
21:06:52filwitAraq: dom96 said there wasn't, but he wasn't 100% sure either
21:06:57Dirksonfowl: That doesn't surprise me much. We're got to release ASAP, and we're generally lone-wolf types... Compile massively limited developer time with a good incentive to code faster, and suddenly alternative languages start looking really nice.
21:07:23fowlfilwit, the parameter might get passed by reference anyways, but not be accessible that way
21:07:24filwitAraq: i would benchmark.. but haven't had luck with that, and don't want to waste a bunch of time on micro-benchmarks right now
21:07:57AraqMFlamer: mind to add your lock free hash table already? I'm nervous to lose this precious work. No offense but people keep disappearing for good here in open source land
21:08:57Dirkson"const" is not equivalent to C's "const", but rather to its #defines, aye? While
21:09:04Araqfilwit: depends on the 'T'; can be a tad slower but you'll have a hard time measuring it
21:09:05DirksonWhile "let" is more equivalent to C's const.
21:09:12fowlDirkson, there are tons of libraries and wrappers ready for game dev, glfw, glfw3, enet, sdl2, sdl, im forgetting a bunch
21:09:23Dirksonfowl: Excellent : )
21:09:24AraqDirkson: you can say that, yes
21:09:31dom96fowl: Dirkson: sfml too.
21:09:37filwitAraq: okay, thanks
21:09:52fowlsfml, chipmunk
21:09:55fowlODE
21:09:59dom96I used it with Nimrod for Ludum Dare and it worked well.
21:10:02fowlassimp
21:10:10fowldevIL
21:10:50MFlamerAraq: Absolutly, I'll submit a pull request now. I dont want it to go to waste either, just have limited free time and many interests. Just talking about it makes me want to work on it, You know how it is I'm sure
21:11:26Dirksondom96: Oh, I bet. I keep watching people do the Ludum Dare and I always think "Boy. I could just about get a new project started in that amount of time..."
21:11:52MFlamerLudum Dara?
21:12:00MFlamerDare
21:12:09MFlamerWhats that?
21:12:12DirksonMFlamer: Make a game in X hours. Usually 3 days I think?
21:12:21DirksonOr maybe 24 hours.
21:12:42MFlamerOh, like a group hack a thon type thing
21:13:09DirksonAye
21:13:16dom9648 hours.
21:13:19DirksonBut each person competes solo to do their game.
21:13:26Hannibal_SmithNo offense but people keep disappearing for good here in open source land <-So true, I also find myself in this situation, I have to close a push request that I opened on another project
21:13:28dom96or longer if you're working in a group
21:13:28Dirksondom96: Thanks. Don't they have ones of varying lengths?
21:14:31DirksonHannibal_Smith: Had a guy spend like three days creating a patch to convert my game to C++... then vanish, leaving nothing behind but a partial patch. People are funny sometimes.
21:15:09MFlamerI must admit I have several half done projects.....
21:15:18Hannibal_SmithDirkson, I did create a patch, be annoyed of programming, make other things for some times
21:15:31dom96Dirkson: They have unofficial ones called MiniLDs too. The official ones are split into the solo and "jam" or whatever they call it where you get more time and you can work with other people.
21:15:31*faassen quit (Quit: Leaving.)
21:15:41Dirksondom96: Neat!
21:16:20dom96I've never actually completed a game in those 48 hours though :\
21:25:56filwitdom96: only cause you didn't have the right tools ;)
21:26:25filwitdom96: making a game with Unity3D in 48 is feasible if your goals are modest enough
21:26:40fowlunity isnt free
21:26:42fowlis it?
21:26:46filwityea it is
21:26:49filwitjust not open
21:26:55fowlis it on linux?
21:26:58filwitand the free version is limited in some ways
21:26:59DirksonI'm not sure many games worth playing are possible in 48 hours.... But then, I tend to be a simulation junkie, myself.
21:27:04filwitnot the editor
21:27:09filwitbut it publishes to linux
21:27:15filwitalthough, rather poorly
21:27:42filwitlol, on Catalyst, Unity3D wont even sense the resolution correctly if you have two monitors
21:28:02filwitwe had to write a hack for it on my computer
21:28:31fowloh lol
21:28:43filwiteven on Windows, if you run the editor in 'Free Aspect Camera' mode (where the camera isn't locked to a specific resolution, then the editor incorrectly picks up my monitors resolution instead of the windows resolution
21:29:27filwitthey're really slow at fixing bugs too.. like, really slow..
21:29:41filwitthey're still using Mono 2.6 for instance..
21:29:51filwitit's got a horrible GC and C# bugs
21:30:00fowlim working on a skeletal sprite animation tool for things like this: http://opengameart.org/content/orc-pieces
21:30:24fowlby working on i mean i started the project a few times and havent been working on recently but im about to give it another go
21:32:24filwitlol, nice
21:44:49*zahary quit (Read error: Connection reset by peer)
21:45:01*zahary joined #nimrod
21:46:10DirksonOoph, definitely not a fan of the procedure syntax. I'm used to using the definition of a function as a simple shorthand to how it's called: Function(type name, type name, type name) . This Procedure(name, name: type; name, name, name: type) thing is not nearly so readable. I guess I could just do Procedure(name: type; name: type; name: type) to correct for that?
21:46:42Araqyeah you can do that
21:46:56Araqbut in C this common:
21:47:04AraqreturnType
21:47:12Araqfoo(bar baz)
21:47:20Araqwith a newline
21:47:30Araqproving you wrong :P
21:47:54DirksonNot sure I understand?
21:48:14AraqC's declaration syntax is not readable otherwise the newline convention wouldn't exist
21:49:18filwitDirkson: i much prefer listing the types/attributes after the var name. that way when you parse the code, you always scan the front of the line to find the symbol your looking for
21:49:54fowlvoid *foo(int x[10]); no thx
21:50:30DirksonOh, I'm not saying either way of doing it is perfectly readable. There's plenty about C I hate, (Pointers. I'm looking at pointers.) and its syntax is frequently one of those things. ; on every line? Seems silly.
21:51:15filwitDirkson: yeah, the whole ';' thing really bugs me. Especially when more modern languages like D and C# copy it..
21:51:22Dirksonfilwit: Behind or in front makes little difference to me - I think either one is pretty easy to get used to.
21:52:12filwitDirkson: well about the (x, y, z:float) thing, it's really nice for things like Vectors (as illustrated)
21:52:25DirksonI also share a quiet love for the indentation-as-control-structure idea, like Nimrod and Python do. Things that force people to write readable code are always a bonus :D
21:52:41Dirksonfilwit: Hey, that's kinda a fair point!
21:53:29Dirksonfilwit: But at the same time, the extra syntax makes it hard to read for the other purpose. I'd need to add comments onto every function reminding myself how to call it.
21:53:57filwitDirkson: yeah, i'm a C-style programmer myself. but the python style is starting to grow on me. even things like shifting lines around is a bit easier when you don't have to worry about brackets
21:54:25DirksonI always describe python as "My favorite language that I never use."
21:54:45filwitDirkson: honestly, the (a, b, c:Foo) thing is pretty easy to read once you get used to it. i remember it being a bit hard to part initially as well
21:54:50DirksonEither something is so trvial that I can use bash for it, or it needs to be cpu-efficient so I can't use python for it.
21:54:53Dirksonfilwit: *nods*
21:57:51*OrionPK quit (Ping timeout: 250 seconds)
22:04:09DirksonI also really like the "discard" thing. I am very guilty of throwing away return values all the time, and a mechanism for me to say "Yes, I really meant to do that" is probably a good idea that will result in better code.
22:04:54DirksonHey! Named arguments are an interesting way to do things. That really promote readability.
22:05:12fowlaraq will be glad to hear that, most ppl complain about discard
22:08:18DirksonThe most frequent reason I throw away return values is because I'm assuming that the return value can only ever be true - I.e. I'm ignoring potential errors. So a mechanism that makes me stop and think about what I'm doing, and consciously sign-off on it, is a good way to ensure I check for a lot more errors :D
22:08:45fowlhowd you find nimrod?
22:09:19DirksonI was looking up benchmarks for languages, and came across one for nimrod. It seems to have really good figures, so I looked into it... And I've been liking what I've been seeing : )
22:09:31Dirkson(With a couple minor exceptions, which is going to be true of anything)
22:10:02filwitDirkson: btw, you can avoid needing to call discard if you add the {.discardable.} pragma to the proc
22:10:21DirksonSee, now I probably shouldn't know that :D
22:10:34filwitDirkson: cannot be unknown
22:10:41DirksonI know, right?
22:10:43filwitlol
22:11:24Araqwhat's the deal? {.discardable.} is verbose so nobody uses it everywhere (except fowl perhaps)
22:12:02filwityou should have named it {.you_can_optionally_call_discard_on_this_proc.}
22:12:05filwitjust in case
22:12:09Hannibal_SmithAhahaha
22:12:15DirksonAraq: Oh, I was saying how much I liked "discard", so they're telling me how to disable it :D
22:12:28dom96You could create a macro which adds it to every proc :P
22:12:53*Araq doesn't mention {.push discardable.}
22:13:00dom96oh yeah
22:13:15filwitdom96: or just {.push:discardable.} include ... {.pop:..}
22:13:42filwitoh damn, Araq beat me to that
22:14:00dom96Slipped my mind hah.
22:14:26dom96And I wanted to give Dirkson an idea of what macros can do.
22:15:15filwitdom96: i really liked Araq's HTML DSL slides on his talk to illustrate Nimrod's template/macro capabilities
22:15:26*OrionPK joined #nimrod
22:15:26filwitbtw, any word on the video?
22:15:49AraqI need to contact them ...
22:16:06dom96I think it will take them a while.
22:16:18filwitplease do, i would really like to see it :)
22:16:24Hannibal_Smithfilwit, can you pass a link to it?
22:16:52filwitHannibal_Smith: don't have the link. Dom96?
22:16:55dom96They're going to wait a couple of months and then start releasing each video every couple of days.
22:17:24dom96Hannibal_Smith: http://dan-ball.jp/en/javagame/dust/
22:17:27dom96argh
22:17:28filwitdom96: that makes sense though i guess. otherwise the tickets to the actual even wouldn't be worth as much
22:17:30dom96fail at copy paste
22:17:38dom96Hannibal_Smith: http://nimrod-code.org/talk01/slides.html
22:17:38filwitlol
22:17:43Hannibal_SmithThank you
22:18:12dom96filwit: yeah
22:18:15filwitHannibal_Smith: slide i'm refering too starts around slide 17
22:18:45Dirkson"Strings in Nimrod are both zero-terminated and have a length field." - How interesting!
22:19:05filwityeah.. i didn't even know that
22:19:14Araqoh yeah we got "this sucks" for that feature too
22:19:19filwitzero terminated to play well with C?
22:19:25Araqyeah
22:19:34Araqand because I love it for parsing
22:19:42DirksonI'm a big fan of length field strings. I implemented them for the network portion of my game.
22:19:43filwiti see
22:20:19Araqyou can be ignorant of it though and things just work
22:20:22DirksonI am... Going to really enjoy not having to deal with C's string library.
22:22:01Dirkson"Array access is always bounds checked" - What behavior results if I try to access an array outside of its area?
22:22:59Hannibal_Smithfilwit, yes it's interesting
22:24:33filwitHannibal_Smith: it illustrates a key power of Nimrod I never realized until recently (even though i've seen syntax like that before). Nimrod is capable of allowing third-parties to create and promote their own DSLs, without really having to do much of any work to actually get the DSL
22:25:17OrionPKdom96 so yeah
22:25:20OrionPKthe error
22:25:21OrionPKError: unhandled exception: A non-blocking socket operation could not be complet
22:25:21OrionPKed immediately.
22:25:33OrionPKworks fine if I just call run()
22:25:35fowlDirkson, there is compiletime check so if you var x: array[10, int]; echo x[20] it can be detected at compile time and runtime checks (crash, stack trace, etc)
22:25:52Dirksonfowl: Excellent
22:26:06dom96OrionPK: can you give me a gist of your code and a traceback?
22:26:18filwitHannibal_Smith: for instance, in my game-engine design, the plan is to build out multiple functions from different "parts" (my macro for a 'class') into one optimized function for release, but build it dynamic (for editing) otherwise. No other language allows me to do that as easially as Nimrod does
22:26:18Araqactually we have the technology to detect it fully at compile-time ...
22:26:30fowlDirkson, i shouldnt say crash i mean exception is thrown
22:26:31OrionPKdom96 i need maybe 20 mins
22:26:53dom96OrionPK: ok
22:27:08Araqgot bored with the flow dependent typing though so it won't be in 0.9.4
22:27:10fowlAraq, var idx = 20; echo x[idx] cant be
22:27:30Araqfowl: yes it can
22:27:58fowlwhat if i put it in a loop where i randomly increase or decrease idx
22:28:03fowldetect that
22:28:07Hannibal_SmithIs there some book that explain some meta programming techniques? Or they are too bond to the language?
22:28:29AraqHannibal_Smith: anything with "Lisp" in its name applies I think
22:28:34Hannibal_SmithOk
22:28:50Araqthough the details are different of course
22:29:07Hannibal_SmithYes at the moment I'm more interested in the concepts
22:29:34Araqfowl: the trick is that the compiler makes you insert 'if idx >= 0 and idx <= high(a): ' if it gets too hard to prove ;-)
22:29:36Hannibal_SmithI only know the basic metaprogramming that C++ offers
22:30:09filwitHannibal_Smith: meta-programming in Nimrod is a bit different than C++
22:30:55*io2 quit ()
22:30:55Hannibal_SmithYes, I'm seeing this from Araq slides
22:31:27OrionPKhttps://gist.github.com/onionhammer/7109327
22:31:28OrionPKbrb
22:32:34fowlAraq, i want to do "if idx in 0 .. high(a):" instead
22:33:27fowlHannibal_Smith, templates are simple substitution based things, macros are compile-time functions that operate on AST, you build it procedurally/imperatively
22:34:02filwitHannibal_Smith: technically, I think Nimrods is much more powerful (and incorporates type-safe generics), but doing things like making a type which changes is structure based on template params feels much different than C++/D. You call a compile-time function which chooses (or generates) a specific type, rather than defining a templated-type
22:35:53dom96OrionPK: Works for me.
22:36:26OrionPKon windows?
22:36:44dom96OrionPK: nah, linux. Must be Windows specific then.
22:36:57OrionPKyeah
22:36:57fowldom96, this dust game is really neat
22:36:59dom96so how about a traceback?
22:37:05OrionPKit does say "OSError"
22:37:17OrionPKadded it to the gist
22:37:28OrionPKoop sorry
22:37:31OrionPKthat's the wrong one
22:37:46OrionPKjust ignore the 1st two lines
22:38:14Dirksonfowl: Which one are you playing? : )
22:38:24fowlthe one dom just linked
22:38:43fowlhttp://dan-ball.jp/en/javagame/dust/
22:38:47dom96fowl: Yeah, it's really addictive.
22:38:56Dirksonfowl: Neat ^^
22:39:08fowlim about to say f modular sprites and just make a game out of little figures
22:39:15fowllittle stick figures
22:39:37fowland particles
22:39:51Dirksonfowl: Get it done, and make the rendering of the stick figured modular. You decide you want to fix it later? Come back and replace the underpinnings : )
22:40:13dom96OrionPK: Yeah... I think httpserver is wrong.
22:40:20OrionPKoh yeah?
22:40:33OrionPKso openSSL is wrong and httpserver is wrong ;)
22:40:47dom96It's what happens when I rush the async implementation.
22:41:10OrionPKnp, I was wondering how much you were using that one vs the run() approach
22:41:29fowlDirkson, these are nice tho, the physics making their legs all wobbly
22:41:37fowli should have went to college
22:41:46fowli would have taken a physics class
22:42:24dom96httpserver needs a rewrite anyway though
22:42:55DirksonSchool can be useful... But it's not required for success. I didn't even graduate highschool.
22:42:55fowldom96, fowltek already exists. Overwrite? [y/N] <- this should be [Y/n]
22:43:17fowlwell i didnt even take physics in HS
22:43:37fowli did the bare minimum
22:43:49dom96fowl: yeah, perhaps you're right. It gets annoying. But overwriting is destructive so having the default be 'yes' seems like a bad idea.
22:44:39Dirkson./koch install seems to assume that nimrod is already in the PATH , when it ought to be the thing putting nimrod in the PATH : )
22:44:52dom96if only I could just quit school and code all day... that would be great.
22:45:28fowlDirkson, dont bother "installing" if you're on linux
22:45:35fowljust make a symlink on your path
22:45:35Hannibal_Smithdom96, you are doing a lot for an high school student
22:45:42Dirksonfowl: All I need is "nimrod"?
22:45:45fowlyea
22:45:49dom96OrionPK: My plan was to make asyncio nicer and then implement a proper async version of these stdlib modules, but I ran out of time... :\
22:45:56Dirksonfowl: Cool. Thanks
22:46:07dom96Hannibal_Smith: Thanks :)
22:46:09MFlamerdom is in high school?
22:46:10DirksonWhat is "CAAS" support?
22:46:18dom96MFlamer: Yeah.
22:46:23fowlyou'll want to get babel too
22:46:30MFlamerNice, your savage
22:46:35dom96Dirkson: Compiler as a service.
22:47:06fowldom96, i have to reinstall fowltek every time i push a change
22:47:15Dirksondom96: *googles* I'm not sure that helps me much :-/ Can you explain more?
22:47:18fowli could probably just do yes | babel install fowltek
22:47:30OrionPKdom96 ah, so far it seems to work *fairly* well
22:47:39dom96Dirkson: It's basically a more efficient way for an IDE to query the compiler for information.
22:47:47OrionPKI've got the irc client bit working with multiple clients on multiple servers through 1 dispatcher
22:47:51Dirksondom96: Oh, cool : )
22:48:04OrionPKI wanted to add a web server into that dispatcher as well, and thats where I got into trouble
22:48:43dom96OrionPK: In that case I will fix it for you as soon as I can. or you can try it yourself if you wish.
22:49:16dom96alternatively you can use scgi + nginx
22:49:20OrionPKI'm not in enough of a hurry to try to fix it myself :p
22:49:27fowllol^
22:49:51OrionPKgot a busy week, so I can afford to hold off on this little side project
22:50:26OrionPKdom96 if you want me to try any patches out before you check them in (and cant reproduce the issue urself), ping me
22:50:53dom96OrionPK: Sure. Can you create an issue on github just as a reminder for me?
22:51:00OrionPKsure
22:51:05*Hannibal_Smith quit (Quit: Sto andando via)
22:51:06dom96thanks
22:53:11OrionPKhttps://github.com/Araq/Nimrod/issues/632
22:53:49Araqfowl: 'in' or '>= and <= ' doesn't matter
22:58:45MFlamerAraq: do you have any hints on how I would fix the generics bug I mentioned last night? Maybe a hunch on a compiler function or file?
22:59:08MFlamerI know your of to sleep soon :-)
22:59:51fowlAraq, are you trying to get rid of runtime detection for arrays completely
23:05:45AraqMFlamer: the type merging is done in semtypinst.nim and I don't know how it works; zahary rewrote it ... :-/
23:05:53reactormonkDirkson, new people \o/
23:05:59Dirksonreactormonk: *waves*
23:06:30DirksonHmm. Having trouble making nimrod go. ./koch install assumes nimrod is already installed, and my attempt to install it myself seems to have resulted in a non-working nimrod.
23:06:32AraqMFlamer: takes a month to understand the concepts, another to understand the algorithm and then you can try to fix it
23:06:58MFlamerOuch...
23:07:27MFlamerAnd Zahary seems busy lately
23:07:48filwitDirkson: download the csources
23:08:02MFlamerMaybe I'll have to start ofering $ prizes for bug fixes
23:08:37Araqdon't worry I'll fix it ... eventually
23:08:47filwitDirkson: https://github.com/nimrod-code/csources
23:09:21fowlDirkson, just make a symlink man
23:09:39DirksonYeah - Symlink works.
23:09:43filwitDirkson: also, what distro are you on?
23:09:48Dirksonfilwit: Debian
23:09:59filwitDirkson: okay, nevermind.
23:10:03MFlamerok, I guess I'll have to finish the table as leverage for favors
23:10:12filwitDirkson: Nimrod is in the Arch main repos, just so you know :)
23:10:17MFlamer;-)
23:10:27Araqfowl: nah, this form of dependent typing gets annoying rather quickly. I will make the compiler emit warnings instead.
23:10:47Dirksonfilwit: Aye : ) Don't often use arch - Good OS, and remarkably stable for what it is... buuut my PC is a workhorse. It does what it does for years on end, and the less I have to touch it the better :D
23:10:49filwitbtw, anyone know if Nimrod is in Manjaro?
23:11:32filwitDirkson: yeah, idk. I've been running Arch stably for over a year now really
23:11:46filwitDirkson: every once in awhile they'll introduce a bug, and i'll report
23:12:20filwitDirkson: but I use a lot of art software, and having that always up-to-date is nice, rather than hunting for obscure, version-specific PPAs
23:12:28Dirksonfilwit: Yup. It's the bugs I avoid by sticking with debian. Bugs are vanishingly rare in testing - And less so when you just sit on an ages-old tested and never update anything unless forced :D
23:12:58Dirksonfilwit: Buuut in return all my software is from 2007.
23:14:15DirksonIs there an easy way to tell nimrod what compiler to use?
23:14:55Araqyeah
23:15:09Dirksonconfig/nimrod.cfg ? Just found that : )
23:18:20dom96good night guys
23:18:36DirksonNight!
23:18:56filwitnight, dom96
23:22:10MFlamerdoes n=<optimized out> in gdb mean that gcc optimized it away?
23:22:21MFlameror, nimrod did?
23:23:07MFlamerI would assume that not much would get optimized away in a debug build
23:23:26Araqgcc did
23:23:33Araqfor whatever reason
23:25:38MFlamerok
23:26:30Araqgood night
23:26:37filwitnight, Araq
23:26:49filwiti'm taking off as well, later folks
23:27:00filwitgood to meet you, Dirkson
23:27:05*filwit quit (Quit: Leaving)
23:27:15DirksonNice to meet you too, person who is now gone :-/
23:27:41MFlamergood night
23:37:27*OrionPKM joined #nimrod
23:51:30OrionPKdom96 im getting an error in OSX as well
23:58:55*OrionPKM quit (Remote host closed the connection)