code logs -> 2013 -> Wed, 17 Apr 2013< code.20130416.log - code.20130418.log >
--- Log opened Wed Apr 17 00:00:08 2013
00:02 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
00:02 mode/#code [+qo Vornicus Vornicus] by ChanServ
00:03 ToxicFrog|W`rkn is now known as ToxicFrog
01:12
<@sshine>
for a bunch of processes that procure a semaphore, is there any guarantee (e.g. in POSIX) that semaphores will be used FIFO or in other ways fairly?
01:14
<@Azash>
Found an rtems mailing list that says it's FIFO blocking
01:14
<@Azash>
Meanwhile someone on StackOverflow says there is no guarantee
01:14
<@sshine>
is that for POSIX semaphores, or semaphores in general?
01:15
<@Azash>
Hm, the latter is C but does not mention POSIX
01:15
<@sshine>
Pthreads, eh?
01:15
<@Azash>
Here's the former http://www.rtems.com/ml/rtems-users/2004/april/msg00086.html
02:13 Turaiel[Offline] is now known as Turaiel
02:32
<&ToxicFrog>
sshine: semaphores are part of a POSIX standard independent of pthreads, actually, hence why they don't have the pthread_ prefix
02:34
<&ToxicFrog>
Anyways, the POSIX standard does not specify how the scheduling works beyond a "may be subject to priority inversion" warning
02:35
<&ToxicFrog>
If you are targeting a specific implementation (e.g. a specific linux kernel version) you can of course come up with guarantees specific to that.
02:58 Turaiel is now known as Turaiel[Offline]
03:11 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
03:15 Kindamoody[zZz] is now known as Kindamoody
03:44 syksleep is now known as Syk
04:06 * McMartin prepares a 2KB patch.
04:06
<&McMartin>
This is much more alarming than that 1.3MB patch from last week.
04:07 * McMartin decides he will hold off on it and trigger it so that it will be done with the automated aspect of testing when he arrives at work tomorrow.
04:07 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
04:20 himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code
04:20 mode/#code [+o himi] by ChanServ
04:25 Turaiel[Offline] is now known as Turaiel
05:09 Kindamoody is now known as Kindamoody|out
05:14 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
05:29 ErikMesoy|sleep is now known as ErikMesoy
05:31 himi [fow035@Nightstar-68d8c119.in-addr.csiro.au] has joined #code
05:31 mode/#code [+o himi] by ChanServ
05:40 Turaiel is now known as Turaiel[Offline]
05:40 Turaiel[Offline] is now known as Turaiel
06:34 Derakon is now known as Derakon[AFK]
07:06 Kindamoody|out is now known as Kindamoody
07:23 Turaiel is now known as Turaiel[Offline]
08:27 celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
09:56 RichyB [richardb@Nightstar-228a334c.plus.com] has joined #code
10:01 Reiver [quassel@Nightstar-3762b576.co.uk] has joined #code
--- Log closed Wed Apr 17 10:41:57 2013
--- Log opened Wed Apr 17 10:42:05 2013
10:42 TheWatcher[afk] [chris@Nightstar-3762b576.co.uk] has joined #code
10:42 Irssi: #code: Total of 37 nicks [20 ops, 0 halfops, 0 voices, 17 normal]
10:42 mode/#code [+o TheWatcher[afk]] by ChanServ
10:42 Irssi: Join to #code was synced in 38 secs
10:43 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [[NS] Quit: Leaving]
11:05 Kindamoody is now known as Kindamoody|out
12:41 himi [fow035@Nightstar-68d8c119.in-addr.csiro.au] has quit [Client closed the connection]
12:56 ToxicFrog is now known as ToxicFrog|W`rkn
13:03 himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code
13:03 mode/#code [+o himi] by ChanServ
13:36 RichyB [richardb@Nightstar-228a334c.plus.com] has quit [Ping timeout: 121 seconds]
14:37 RichyB [richardb@Nightstar-228a334c.plus.com] has joined #code
14:56
<@froztbyte>
http://www.pcworld.com/article/2033318/black-annex-is-the-best-qbasic-game-youve -ever-seen.html
14:58
< RichyB>
I find the "it requires a 2.6GHz CPU" a bit surprising.
14:59
< RichyB>
There are some open-source BASIC implementations which go really fast. The language doesn't have (many) semantics that are hard to turn into very terse, quick object code, after all.
15:56
<@froztbyte>
yeah, I'm left to wonder how much time the author had to spend building some advanced constructs
15:56
<@froztbyte>
I mean, if all you get are the bricks, you can still get pretty far just with those
--- Log closed Wed Apr 17 16:11:13 2013
--- Log opened Wed Apr 17 16:11:20 2013
16:11 TheWatcher[afk] [chris@Nightstar-3762b576.co.uk] has joined #code
16:11 Irssi: #code: Total of 36 nicks [20 ops, 0 halfops, 0 voices, 16 normal]
16:11 mode/#code [+o TheWatcher[afk]] by ChanServ
16:11 Irssi: Join to #code was synced in 38 secs
16:31 Turaiel[Offline] is now known as Turaiel
16:38 Turaiel is now known as Turaiel[Offline]
16:41
< RichyB>
Maybe it's running in actual QBASIC.
16:41
< RichyB>
Y'know, actual old steam-powered-interpreter QBASIC.
16:42
<@froztbyte>
mmmmmmmm
16:42
<@froztbyte>
that'd be tricky
16:42
<@froztbyte>
I expect the very first issue you'd have is translating the compiler output
16:42
<@froztbyte>
which, you know, you could actually beat into shape (by further abuse of qbasic)
16:43
< RichyB>
Huh? All he'd have to do is ship QBASIC.EXE along with his BAS file(s).
16:43
< RichyB>
(possibly with a copy of DOSbox)
16:43
<@froztbyte>
yes, that's my point, *old* qbasic probably won't run on modern platforms
16:44
<&ToxicFrog|W`rkn>
It definitely won't, but I don't know what you're on about wrt "translating the compiler output"
16:44
<@froztbyte>
and (automatically correctly) surrounding shit with dosbox for the correct visual mode switches and stuff is probably another world of pain
16:44
<&ToxicFrog|W`rkn>
Old QBASIC is a DOS IDE and interpreter.
16:44
<@froztbyte>
ToxicFrog|W`rkn: well, I never used qbasic
16:44
<@froztbyte>
I'd assumed it was a compiled language
16:44
<&ToxicFrog|W`rkn>
You write your program (in the IDE or a separate editor) and load and run it from inside QBASIC.EXE
16:44
<@froztbyte>
I'm only familiar with the general state of things around that timeline, have used very few of them personally
16:45
<&ToxicFrog|W`rkn>
Wrapping it with dosbox is just a matter of putting dosbox itself in the same dir along with a dosbox.conf with an appropriate [autoexec] section; it's not actually very hard at all.
16:45
<&ToxicFrog|W`rkn>
It does, of course, mean that you need a lot more beef to run it,.
16:46
<@froztbyte>
last time I tried doing anything with dosbox, I was reminded of 1994 and getting highmem.sys and everything else playing along well
16:46
<&ToxicFrog|W`rkn>
Yeah, that hasn't really been the case for, at minimum, 3+ years
16:46
<@froztbyte>
ah
16:46
<@froztbyte>
sweet
16:46
<&ToxicFrog|W`rkn>
Longer if you count the first usable frontends (DBGL/D-FEND)
16:47
<&ToxicFrog|W`rkn>
Seriously, if you're configuring dosbox yourself, just use DBGL.
16:47
< RichyB>
QBASIC.EXE is an interpreter.
16:47
< RichyB>
I think you'd just run it as something like "QBASIC.EXE /run DONKEY.BAS"
16:47
<&ToxicFrog|W`rkn>
If you're releasing something wrapped in DOSBOX, you use DBGL, then package up dosbox + your program + the config file DBGL generated and there you go.
16:47
< RichyB>
? put that in dosbox's [autoexec] section and you have yourself a fully-packaged DOS game.
16:47
< RichyB>
What TF said.
16:48
<&ToxicFrog|W`rkn>
Or you can just configure it by hand; set the graphic options as you see fit and then just [autoexec] mount game c; c:; qbasic /run donkey.bas
16:48
<&ToxicFrog|W`rkn>
(this is in fact how GOG does all of their DOS releases, modulo some additional trickery to get CD music and whatnot working)
17:19 Syk is now known as syksleep
18:06 Derakon [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code
18:06 mode/#code [+ao Derakon Derakon] by ChanServ
18:06 * Derakon mutters at transformation matrices.
18:06
<&Derakon>
I must be misunderstanding something.
18:06
<&Derakon>
Say I want to apply three rotation transformations to an array. So I have their transformations A, B, and C.
18:07
<&Derakon>
I start with an identity matrix to represent "no transformation". Then I do C*B*A*I to get the transformation matrix.
18:07
<&Derakon>
But the problem is that A*I is losing data somehow.
18:08
<&Derakon>
For example: http://pastebin.com/hZxvvDV7
18:09
<&Derakon>
(That's a 90-degree rotation about the Z axis there)
18:40 celticminstrel [celticminst@Nightstar-e83b3651.cable.rogers.com] has joined #code
18:40 mode/#code [+o celticminstrel] by ChanServ
19:04 Turaiel[Offline] is now known as Turaiel
19:11
<&Derakon>
Turns out the problem was that numpy was doing per-element multiplication instead of matrix multiplication; I needed to use numpy.dot() instead. Whoops.
19:12
< RichyB>
AH
19:13
< RichyB>
(*) means element-wise multiplication if you're using numpy.array()s but matrix-multiplication if you're using numpy.matrix()
19:13
< RichyB>
Man, that's not even slightly confusing.
19:13
< RichyB>
Oh wait yes it is. ;P
19:14
< RichyB>
Sew, use array() everywhere and use .dot() to get matrix multiplication.
19:27
<&Derakon>
Right.
19:43 * RichyB was seriously confused there too.
19:43
< RichyB>
A by I_A should always be equal to A, no matter what A is.
20:06 Kindamoody|out is now known as Kindamoody
20:06 Turaiel is now known as Turaiel[Offline]
20:44 Kindamoody is now known as Kindamoody[zZz]
21:01 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
21:01 mode/#code [+qo Vornicus Vornicus] by ChanServ
21:40 ErikMesoy is now known as ErikMesoy|sleep
23:09 d4de [olorin@687AAB.418D22.95850D.D4B73D] has quit [Ping timeout: 121 seconds]
23:10 d4de [olorin@687AAB.418D22.06F82B.900D20] has joined #code
23:17 Turaiel[Offline] is now known as Turaiel
23:21 Derakon [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [[NS] Quit: leaving]
23:38 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code
--- Log closed Thu Apr 18 00:00:22 2013
code logs -> 2013 -> Wed, 17 Apr 2013< code.20130416.log - code.20130418.log >

[ Latest log file ]