code logs -> 2013 -> Thu, 10 Jan 2013< code.20130109.log - code.20130111.log >
--- Log opened Thu Jan 10 00:00:10 2013
00:04
<@simon`>
chatbots are only true once they try to deny the fact that they're chatbots.
00:05
<@simon`>
(intelligent) humans acknowledge the impossibility to clearly distinguish.
00:48 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
00:49 mode/#code [+qo Vornicus Vornicus] by ChanServ
01:06 syksleep is now known as Syk
01:07 Thalass|NB [thalass@Nightstar-724ec5eb.bigpond.net.au] has quit [Ping timeout: 121 seconds]
01:08
< Syk>
i had someone in my year 11 class who i swear was operating on Markov chains
01:08
< Syk>
and that he'd only been fed sydney party language and Dane Cook shows
01:10
<&McMartin>
ha ha
01:11 You're now known as TheWatcher[T-2]
01:11
< Syk>
"The problem was that Watson couldn't distinguish between polite language and swearing. Apparently it picked up some bad habits from reading Wikipedia and started using terms like "bullshit" in an answer to a researcher's query."
01:11
< Syk>
ahahah what
01:12
< Syk>
http://www.fudzilla.com/home/item/30064-ibm-accidentally-taught-watson-to-swear
01:18 You're now known as TheWatcher[zZzZ]
01:34 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code
01:34 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has quit [Connection closed]
01:42
<@Alek>
ahaha a step forward for AIs
01:55 Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds]
01:55 Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code
01:55 mode/#code [+ao Derakon Derakon] by ChanServ
02:08 Derakon is now known as Derakon[AFK]
02:30 Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has quit [Ping timeout: 121 seconds]
02:48 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code
03:06 Kindamoody[zZz] is now known as Kindamoody
04:26 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: Program Shutting down]
04:44 Derakon[AFK] is now known as Derakon
05:16 io|bleh is now known as iospacedout
06:04 Syk_ [the@Nightstar-b9cf01f7.iinet.net.au] has joined #code
06:05 Syk_ [the@Nightstar-b9cf01f7.iinet.net.au] has quit [[NS] Quit: leaving]
06:05 Syka [the@Nightstar-b9cf01f7.iinet.net.au] has joined #code
06:06 Syk [the@Nightstar-df48a937.iinet.net.au] has quit [Ping timeout: 121 seconds]
06:15 d4de [olorin@687AAB.1E386D.9FE338.92E600] has joined #code
06:19 d4de [olorin@687AAB.1E386D.9FE338.92E600] has left #code []
06:19 d4de [olorin@687AAB.1E386D.9FE338.92E600] has joined #code
06:32 Derakon is now known as Derakon[AFK]
06:40 ErikMesoy|sleep is now known as ErikMesoy
06:42 Syka_ [the@Nightstar-93f02c9a.iinet.net.au] has joined #code
06:43 Syka [the@Nightstar-b9cf01f7.iinet.net.au] has quit [Ping timeout: 121 seconds]
07:11 * Tarinaky glares
07:11
<@Tarinaky>
value += numpy.dot(tl, (x % period_x, y % period_y))
07:11
<@Tarinaky>
Undefined variable from import: dot
07:11
<@Tarinaky>
Really? Because help(numpy.dot) indicates it's right there.
07:13
<@Tarinaky>
Why is PyDev not finding it. The interpretter finds it..
07:35
<@Tarinaky>
Wait, hang on.
07:35
<@Tarinaky>
Why the fuck is numpy.dot returning a non-scalar?
07:35
<@Tarinaky>
o.o
07:39 * Tarinaky arghs.
07:59
<@Tarinaky>
It'd also be nice if I had an internet connection.
08:00 Tarinaky_mibbit [0264a30b@Nightstar-36f67fd0.mibbit.com] has joined #code
08:00
<@Tarinaky>
Oh, now it works again.
08:00 Tarinaky_mibbit [0264a30b@Nightstar-36f67fd0.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
08:00
<@Tarinaky>
"TypeError: 'float' object is not iterable"
08:01
<@Tarinaky>
I'm not /iterating/
08:01
<@Tarinaky>
Oh hang on.
08:02 * Tarinaky facepalms.
08:13
<@Tarinaky>
Oh for fuck's sake why is nothing working this fucking morning.
08:15 harlow [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code
08:16 mac [mac@Nightstar-fe8a1f12.il.comcast.net] has quit [Ping timeout: 121 seconds]
08:35 Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has quit [[NS] Quit: shutting down!]
08:44 Syka_ [the@Nightstar-93f02c9a.iinet.net.au] has quit [Ping timeout: 121 seconds]
08:45 Syka [the@Nightstar-93ec8e0b.iinet.net.au] has joined #code
08:47 * Vornicus <3 his klotski solver.
08:48
<~Vornicus>
though I do need to teach it to use data files.
08:49
<~Vornicus>
...and handle multiple key objects.
08:56 Syka_ [the@Nightstar-91df8d39.iinet.net.au] has joined #code
08:59 Syka [the@Nightstar-93ec8e0b.iinet.net.au] has quit [Ping timeout: 121 seconds]
09:01 Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has joined #code
09:07 Kindamoody is now known as Kindamoody|out
09:51 Syka_ [the@Nightstar-91df8d39.iinet.net.au] has quit [[NS] Quit: leaving]
09:51 Syka [the@Nightstar-91df8d39.iinet.net.au] has joined #code
10:19 You're now known as TheWatcher
10:32 harlow [mac@Nightstar-fe8a1f12.il.comcast.net] has left #code ["Leaving"]
10:34 RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has joined #code
10:45 * Vornicus throws his klotski solver at "The Diabolical Box", the final puzzle in the Layton game of the same name. 90 seconds, 160k states, 78 moves to the solution.
11:38 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [Client closed the connection]
11:44
<@Tarinaky>
Not sure if there's something wrong with my code but my perlin noise generator is slow.
11:47
< Syka>
isn't perlin noise slow
11:52 iospacedout is now known as iospace
11:53
< RichyB>
Not atrociously, IIRC it's supposed to take like 20 adds+multiplies to generate one value from it. Perlin invented the algorithm on rendering clusters that were being used to draw the original Tron, way back when. You can put Perlin in a fragment shader and draw it across the whole monitor on pretty much any desktop machine.
11:54
<@Tarinaky>
Struggling to convert the list my perlin noise outputs into something pyglet.image can use.
11:57
<@Tarinaky>
It'd help if there was a list of valid format strings.
11:57
<@Tarinaky>
So I could figure out what a greyscale image is.
12:05
< RichyB>
I don't know the answer to that but it'll probably be in OpenGL-related docs.
12:06
<@Tarinaky>
Okay. Why the hell am I even getting things outside of range :/
12:08
< RichyB>
http://www.pyglet.org/doc/programming_guide/accessing_or_providing_pixel_data.ht ml
12:11
< RichyB>
so to make a greyscale imagedata where each pixel is 1 byte, I think you should be doing something like, imagedata.set_data('L', width, data)
12:29
<@Tarinaky>
I have two other more pressing bugs...
12:29
<@Tarinaky>
1) My perlin noise function is producing values outside the range it should be (ie >255...)
12:29
<@Tarinaky>
2) The blitting is retarded.
12:36
< RichyB>
2) can be caused by 1).
12:37
< RichyB>
e.g. you get a 256 in there, pyglet turns it into a zero - random black pixels amidst the brightest whites.
12:40 Syka_ [the@Nightstar-fb197c7a.iinet.net.au] has joined #code
12:42 Syka [the@Nightstar-91df8d39.iinet.net.au] has quit [Ping timeout: 121 seconds]
12:45 Syka [the@Nightstar-e6951529.iinet.net.au] has joined #code
12:45
<@Tarinaky>
How the merry fuck am I getting these numberds :/
12:45 Syka_ [the@Nightstar-fb197c7a.iinet.net.au] has quit [Ping timeout: 121 seconds]
12:45 Tarinaky_mibbit [0264a30b@Nightstar-4fab16c5.mibbit.com] has joined #code
12:46
< Tarinaky_mibbit>
It'd also be nice if I had a fucking internet connection.
12:46
< Tarinaky_mibbit>
-.-
12:46
< Tarinaky_mibbit>
12:29 <@Tarinaky> 2) The blitting is retarded. is the last thing I saw.
12:51
<@Tarinaky>
Anyway,OH FOR FUCK'S SAKE FUCKING INTERNET CONNECTION PLEASE LET ME USE FUCKING SSH
12:51 Pandemic [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code
12:51 mode/#code [+o Pandemic] by ChanServ
12:51
<@Tarinaky>
Is that too much to ask?
13:08
< Syka>
Tarinaky: yes
13:08
< Syka>
:P
13:19
< Tarinaky_mibbit>
How the hell am I multiplying 1 and 1 and getting 1.38325...
13:21
<@TheWatcher>
Large values of 1
13:21
<@Tarinaky>
1.38325... = dot((0.5444775560244061, 0.8387754115290278), (1-0,1-0))
13:22
<@Tarinaky>
Where dot is defined as def dot ((x0,y0), (x1,y1)):
13:22
<@Tarinaky>
return x0*x1 + y0*y1
13:25
<@TheWatcher>
uh
13:26
<@TheWatcher>
0.5444775560244061 * 1 + 0.8387754115290278 * 1 does equal 1.38325
13:27
<@Tarinaky>
Oh >.<
13:30 rms is now known as RobinStamer
13:31 thalass [thalass@Nightstar-724ec5eb.bigpond.net.au] has quit [[NS] Quit: Leaving]
13:32
<@Tarinaky>
Well I am fucked if I know what the problem is then :/
13:37
<@Tarinaky>
http://pastebin.com/SeQ9dV6Q :/
13:40
<@Tarinaky>
Oh wait.
13:40
<@Tarinaky>
I see ther issue
13:53
<@Tarinaky>
Okay... next issue...
13:53
<@Tarinaky>
I'm not getting a lot of variance.
13:53
<@Tarinaky>
Also, shouldn't it be 0 at the corners?
14:04
<@Tarinaky>
http://3.bp.blogspot.com/-Qvjnt7TKCcU/UO7JxKUulGI/AAAAAAAAALI/N9Z1nBxdJ4Y/s1600/ Untitled.png Yay?
14:27 Tarinaky_mibbit [0264a30b@Nightstar-4fab16c5.mibbit.com] has quit [[NS] Quit: http://www.mibbit.com ajax IRC Client]
14:49
< RichyB>
Yaaaay.
14:51
< RichyB>
Tarinaky: (0.54, 0.84) is a unit vector, but (1, 1) isn't. It should be (0.707, 0.707) instead if the maths you're working from assumes unit vectors.
14:52
<@Tarinaky>
Honestly: Perlin Noise is hard and there's very little in the way of good pseudocode for it >.<
14:52
<@Tarinaky>
I /think/ it's working now (see picture)
14:53
< RichyB>
It looks like it's the right kind of continuous. :|
14:54
<@Tarinaky>
"3! 3! 3!" "Are you sure that's random?" "That's the problem with random numbers."
14:56
< RichyB>
You might be able to find some Perlin noise implementation in GLSL under an acceptable open-source license.
14:57
<@Tarinaky>
I don't really want it for textures though.
14:57
< RichyB>
https://github.com/ashima/webgl-noise/wiki <- sounds good, is MIT license (which is basically identical to 3-clause BSD)
14:57
<@Tarinaky>
Turning it into a texture is just for the benefit of testing.
14:58
<@Tarinaky>
Exploratory programming?
14:58
< RichyB>
Makes sense.
14:58
< RichyB>
You *could* generate it in a texture and then read the texture contents into a buffer, if that doesn't feel too silly. :)
14:59
<@Tarinaky>
It sounds a little like premature optimisation.
14:59
<@Tarinaky>
Doing a calculation on a GPU only to read back into main memory :p
14:59
< RichyB>
It'd be massively premature if you were doing it as an optimisation, rather than just because you happen to have open-source GLSL code lying around. :)
15:01
<@Tarinaky>
Also GLSL opens another can of worms :p
15:01
< RichyB>
Are you just using OpenGL 1.x, then?
15:02
<@Tarinaky>
No, but I'm not using straight OpenGL and I don't know OpenGL of any flavour.
15:02
< RichyB>
pyglet doesn't really overcomplicate things.
15:04 RichyB [richardb@Nightstar-3b2c2db2.bethere.co.uk] has left #code [">:3 This is BunThulhu. Copy him into your part message to help him take over the Internet."]
15:42 * TheWatcher hairpulls, realises the entire structure of this code is wrong, rips it asunder
15:53
< ErikMesoy>
Is this where you use the refactor tractor?
15:59
<@TheWatcher>
More like an entire fleet of combine harvesters in this case
16:05
<&McMartin>
Watch out for Code Gordon Freeman!
16:06 * gnolam wonders what a combinatoric harvester would look like.
16:10
<@froztbyte>
it could have a sign on it, ? with subtle infinity wings on the left and right of it
16:10
<@froztbyte>
like, the circle of the xor sign would form the one part of the infinity sign
16:11
<@froztbyte>
and then you could play with colours for either side and fade through and such
16:11
<@froztbyte>
maybe make the plus out of daggers
16:11
<@froztbyte>
(two infinities! because that's countable ;P)
16:23 Derakon [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code
16:23 mode/#code [+ao Derakon Derakon] by ChanServ
16:24
<&Derakon>
We have a bit of hardware that we control via Telnet.
16:24
<&Derakon>
Unfortunately if our program crashes (which happens moderately often in development), then the connection may not get closed correctly on the other end.
16:24
<&Derakon>
At which point the hardware becomes completely inaccessible until you power-cycle it.
16:25
<&Derakon>
(It never times out and will refuse any other connections)
16:25
<&ToxicFrog>
Awesome.
16:25
<&ToxicFrog>
Any way to fix this on the device's end?
16:25
<&Derakon>
So I'm writing a program whose sole function is to act as a proxy for the device; that program will never crash and the actual client program can talk to it instead of the hardware.
16:25
<&ToxicFrog>
If not I suspect the answer is "connect via a telnet proxy that-" yes.
16:25
<&Derakon>
AFAICT no, though I've sent an email to support.
16:27
<@TheWatcher>
And why on earth telnet?
16:27
<&Derakon>
It's just a dumb device.
16:28
<&Derakon>
I guess the manufacturers didn't want to bother telling their users how to deal with SSH.
16:28
<&Derakon>
(There's also RS-232 though)
16:29
<@froztbyte>
telnet's pretty big still
16:30
<@froztbyte>
especially in the network kit world :(
17:02
<&Derakon>
Whelp, proxy's up and running. That wasn't very hard.
17:02
<&Derakon>
<3 the Pyro library for Python remote object proxys.
17:03
<&Derakon>
The client-side code didn't have to change except that instead of doing "self.zConnection = telnetlib.Telnet(ipAddress, port)" it does "self.zConnection = Pyro4.Proxy('PYRO:PIZProxy@%s:%d' % (ipAddress, port))"
17:04
<&Derakon>
From then on I can treat the object as if it were a Telnet instance.
17:14
<@froztbyte>
zConnection? as in zope?
17:14
<&McMartin>
It is z Connection!
17:14
<&McMartin>
As in zMissiles.fire()
17:14
<&Derakon>
Heh.
17:15
<&Derakon>
I'm not familiar with zope, but no, this is the Z axis motion controller.
17:15
<&Derakon>
Similarly there's self.xyConnection.
17:17
<&ToxicFrog>
It's pronounced "zed", filthy colonials~
17:19
<&Derakon>
That makes me think I'm creating a connection to zombies, when that's precisely the problem I'm trying to work around~
17:24
<@gnolam>
http://arstechnica.com/security/2013/01/critical-java-zero-day-bug-is-being-mass ively-exploited-in-the-wild/
17:25
<&Derakon>
Yeah, I turned Java off in my browser something like ten years ago and never looked back~
17:26
<&McMartin>
p. much
17:27
<@gnolam>
Yeah.
17:27
<@gnolam>
I did so because of "arrrrrrgh java applet *browser freeze for half a minute*" and not security though.
17:29
<@froztbyte>
McMartin: hahahahaha
17:29
<@froztbyte>
Derakon: ah k
17:30 Syka is now known as syksleep
17:32
<&McMartin>
IcedTea keeps worming its way into the browser, and I have to keep re-disabling it >_<
17:33
<@froztbyte>
blacklist the browser module package?
17:33
<@froztbyte>
but yeah, I'd love it if those things could be a bit less implicit
17:33
<@froztbyte>
every now and then I have to fix up application bindings in my browser
18:08 Kindamoody|out is now known as Kindamoody
18:26
<&McMartin>
Also, arghlebargle
18:26 * McMartin is full of insights that are completely useless
18:26
<&McMartin>
Most recently: a significant optimization to Bresenham's line algorithm for the jacked-up memory organization of the C64
18:27
<&Derakon>
Now, now, that's not completely useless!
18:27
<&Derakon>
Just mostly.
18:28
<&McMartin>
Heh
18:29 * McMartin might throw together a little sketch program to test out its correctness before actually going and editing his bitmap libraries
18:30
<&McMartin>
If you haven't experienced the hilarity of C64 hi-res mode, it is basically programmable tiles, but with an 8k chunk of memory so each 8x8 block gets a unique tile instead of the usual 2k block (better known as 'the character set or font') + 1k of tile location data (better known as 'the contents of screen memory')
18:30
<&McMartin>
That means you can blit letters straight from your character ROM to the bitmap, which is kind of nice
18:31
<&McMartin>
But it also means that you have to perform a bitmask operation on (y div 8) * 320 + (x and 0xF0) + (y mod 8) to set a pixel.
18:32
<&McMartin>
The optimization is to only compute this once and then rotate a bitmask through the carry bit across locations to do a sort of "rolling pen"
18:33
<&McMartin>
(You know it's time to change addresses because the 9-bit byte rotation rotated your draw point into the carry flag!)
18:37 Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has joined #code
18:37 Derakon [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [[NS] Quit: Lost terminal]
19:09 Kindamoody is now known as Kindamoody[zZz]
19:16 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has quit [Ping timeout: 121 seconds]
19:20 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has joined #code
19:20 mode/#code [+o Rhamphoryncus] by ChanServ
19:39
< ErikMesoy>
How does Python handle conditional and/or user-initiated import statements? (Checking some assumptions before asking potentially wrong questions. Underlying issue: I want to reconcile "don't slurp" gradual/conditional loading of data with the practice of keeping data in Python objects/classes rather than text files.)
19:44
<&ToxicFrog>
The import doesn't happen until the statement is evaluated.
19:44
<&ToxicFrog>
>>> if False:
19:44
<&ToxicFrog>
... import asdf
19:44
<&ToxicFrog>
...
19:44
<&ToxicFrog>
>>>
19:45
<@iospace>
"term does not evaluate to a function taking 302 arguments" LOLWHAT
19:45 syksleep_ [the@Nightstar-66ec159e.iinet.net.au] has joined #code
19:47 syksleep [the@Nightstar-e6951529.iinet.net.au] has quit [Ping timeout: 121 seconds]
20:13 Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has quit [[NS] Quit: ]
20:17 Alek [omegaboot@Nightstar-56dbba0f.in.comcast.net] has joined #code
20:17 mode/#code [+o Alek] by ChanServ
20:21 syksleep [the@Nightstar-13a9c849.iinet.net.au] has joined #code
20:22 syksleep_ [the@Nightstar-66ec159e.iinet.net.au] has quit [Ping timeout: 121 seconds]
20:34 Vornicus [vorn@Nightstar-221158c7.sd.cox.net] has joined #code
20:34 mode/#code [+qo Vornicus Vornicus] by ChanServ
20:41
<&McMartin>
iospace: Heeee
20:42 syksleep_ [the@Nightstar-d8b68292.iinet.net.au] has joined #code
20:43 ReivDriod [Reiver@Nightstar-801a2c6a.vf.net.nz] has quit [Ping timeout: 121 seconds]
20:43 syksleep [the@Nightstar-13a9c849.iinet.net.au] has quit [Ping timeout: 121 seconds]
20:48
<~Vornicus>
ErikMesoy: a snippet from bisect from python 2.7:try:
20:48
<~Vornicus>
from _bisect import bisect_right, bisect_left, insort_left, insort_right, insort, bisect
20:48
<~Vornicus>
except ImportError:
20:48
<~Vornicus>
pass
20:49
<~Vornicus>
erp, that wasn't the manual linebreak button
20:50
< ErikMesoy>
So basically I can go ahead and "import FinalDungeon" at run-time and late in the game, it seems.
20:50
< ErikMesoy>
Sounds simple
20:51
<~Vornicus>
Absolutely. Whether you want to I don't know.
20:51
<~Vornicus>
McM: now I'm imagining that the missiles that General Tor shoots from the far background are called zMissiles
20:52
< ErikMesoy>
Well, I don't want to in the foreseeable future. :p
20:52
< ErikMesoy>
This is still a text game, so the performance penalty for importing absolutely everything on startup would be trivial. But I'm trying to design on principle, because the idea here is to learn how to do things, not just to have a game.
21:06
<~Vornicus>
imports should only bring in code regardless.
21:06
<~Vornicus>
And code isn't (generally) that much.
21:14 You're now known as TheWatcher[afk]
21:16
<@Rhamphoryncus>
Unless your design also involves unloading needed sections I wouldn't bother deferring the load until later
21:20 Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has quit [[NS] Quit: ]
21:22 Derakon_ [chriswei@Nightstar-a3b183ae.ca.comcast.net] has quit [[NS] Quit: Lost terminal]
21:29 Attilla [Obsolete@Nightstar-9e7fa2b2.range86-162.btcentralplus.com] has joined #code
22:13
<&McMartin>
Oh jeez, this thing
22:13
<&McMartin>
Man, I haven't seen this construct in ages
22:33 ErikMesoy is now known as ErikMesoy|sleep
22:42 syksleep [the@Nightstar-5a892523.iinet.net.au] has joined #code
22:45 syksleep_ [the@Nightstar-d8b68292.iinet.net.au] has quit [Ping timeout: 121 seconds]
22:56 syksleep [the@Nightstar-5a892523.iinet.net.au] has quit [Ping timeout: 121 seconds]
23:02 syksleep [the@A6D346.25B8B8.A2EDC3.F995F6] has joined #code
23:18 harlow [mac@Nightstar-fe8a1f12.il.comcast.net] has joined #code
23:33 You're now known as TheWatcher
23:41 VirusJTG [VirusJTG@Nightstar-09c31e7a.sta.comporium.net] has joined #code
--- Log closed Fri Jan 11 00:00:31 2013
code logs -> 2013 -> Thu, 10 Jan 2013< code.20130109.log - code.20130111.log >

[ Latest log file ]