code logs -> 2010 -> Mon, 25 Oct 2010< code.20101024.log - code.20101026.log >
--- Log opened Mon Oct 25 00:00:05 2010
00:25 You're now known as TheWatcher[T-2]
00:27 You're now known as TheWatcher[zZzZ]
00:47 Attilla [Some.Dude@Nightstar-ee3a5014.threembb.co.uk] has quit [[NS] Quit: ]
01:10 AnnoDomini [annodomini@Nightstar-d04f94af.adsl.tpnet.pl] has quit [Connection closed]
01:29 kwsn is now known as kwsn\packers
01:53 jeroid [jerith@687AAB.5E3E50.ABB2AB.182264] has joined #code
02:04 jeroid_ [jerith@Nightstar-2c9a03ca.wa.co.za] has joined #code
02:07 jeroid [jerith@687AAB.5E3E50.ABB2AB.182264] has quit [Ping timeout: 121 seconds]
02:22 jeroid_ [jerith@Nightstar-2c9a03ca.wa.co.za] has quit [[NS] Quit: Bye]
02:50 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?]
03:06 Vornicus-Latens is now known as Vornicus
03:21
<@ToxicFrog>
Eeeeeeeeee
03:21 * ToxicFrog bounces
03:21
<@ToxicFrog>
I have a proof copy of " Hardware Implementations of an Efficient Packet Classification Algorithm with an Incremental Update Capability"!
03:21
< celticminstrel>
I have no idea what that means, but yay anyway!
03:23
< Tarinaky_>
I've heard of it.
03:23
< Tarinaky_>
I know I have.
03:24
< Tarinaky_>
But I can't remember what it is.
03:24
< Tarinaky_>
Google is non forthcoming at jogging my memory.
03:24
< Tarinaky_>
>.<
03:25
< celticminstrel>
Sometimes that happens.
03:25
< Tarinaky_>
TF: Can you give us a clue?
03:27
<@ToxicFrog>
It's a paper on a new packet classification algorithm, including a comparison of pure software, hardware/software codesign, and pure hardware implementations.
03:28
<@ToxicFrog>
It's ! for me because I implemented the hardware version last year.
04:16
<@ToxicFrog>
So, this is my first credit on a paper in my field.
04:17
< Tarinaky_>
Aahhh. That'll be why I've heard of it then.
04:22
< Namegduf>
Anyone know any good libraries for working with ARC4 in C?
04:22<~Reiver> Congrats, TF! :D
04:32 kwsn\packers is now known as kwsn
04:53 McMartin_ [mcmartin@Nightstar-3600f4e7.pltn13.sbcglobal.net] has joined #code
04:56 McMartin [mcmartin@Nightstar-b7b7b084.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds]
04:57 McMartin_ [mcmartin@Nightstar-3600f4e7.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds]
05:04 Tarinaky_ [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Client closed the connection]
05:04 McMartin [mcmartin@Nightstar-c5470041.pltn13.sbcglobal.net] has joined #code
05:04 mode/#code [+o McMartin] by Reiver
05:26
< celticminstrel>
...wait. Python's documentation for the regular expression syntax actually lists \b as having two completely incompatible meanings! XD
05:35
<@Derakon>
Maybe the interpretation used is context-sensitive?
05:36
< celticminstrel>
I don't know how that could work. I'm guessing the interpretation is actually always "word boundary" and never "backspace character"; the former is what one would generally want, anyway, and backspace could be obtained with \x if it were really needed.
06:09 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
06:15 Stalker [Z@2C3C9C.B2A300.F245DE.859909] has quit [Ping timeout: 121 seconds]
06:18 Stalker [Z@2C3C9C.B2A300.F245DE.859909] has joined #code
06:22 Stalker [Z@2C3C9C.B2A300.F245DE.859909] has quit [Ping timeout: 121 seconds]
06:32 Derakon is now known as Derakon[AFK]
07:49 Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
08:45 Anno[Laptop] [annodomini@Nightstar-8ddde7e1.adsl.tpnet.pl] has joined #code
09:00 Vornicus is now known as Vornicus-Latens
09:10 Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has joined #code
09:22 kwsn is now known as kw-sleep-n
09:33 You're now known as TheWatcher
10:03 kw-sleep-n [kwsn@Nightstar-1def9d24.dyn.centurytel.net] has quit [Ping timeout: 121 seconds]
10:11 Anno[Laptop] [annodomini@Nightstar-8ddde7e1.adsl.tpnet.pl] has quit [[NS] Quit: leaving]
10:41 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code
11:14 Anno[Laptop] [annodomini@F67919.F326B3.98D923.BDA7B6] has joined #code
11:41 Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds]
12:08
< gnolam>
http://codebutler.com/firesheep
12:31
<@TheWatcher>
hm. Wonder if that's work over wired connections
12:31
<@TheWatcher>
*that'd
12:33
< Anno[Laptop]>
If I were on Windows, I'd try it right now.
12:41
< gnolam>
TheWatcher: you'd only have a chance on an unswitched network.
12:43
< gnolam>
But anyone who's ever fired up Wireshark on an open WiFi can tell you that people will send anything over an unsecured network.
12:44
< simon_>
yeah.
12:45
< simon_>
I scared the shit out of my ex-girlfriend's roomies once.
12:45
< simon_>
they ran an open wifi, and so did a whole lot of people in their building.
12:51 Anno[Laptop] [annodomini@F67919.F326B3.98D923.BDA7B6] has quit [[NS] Quit: I must flee.]
13:27
< gnolam>
...
13:27 * gnolam figures out the HL2 model error.
13:28
< gnolam>
Apparently, the filename was too long.
13:32 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
13:35
<@TheWatcher>
...
13:35
<@TheWatcher>
wat
13:39 kw-sleep-n [kwsn@Nightstar-1def9d24.dyn.centurytel.net] has joined #code
13:41
< gnolam>
Pretty much my reaction. :P
13:41 kw-sleep-n is now known as kwsn
13:49 Anno[Laptop] [annodomini@Nightstar-8ddde7e1.adsl.tpnet.pl] has joined #code
14:45
< gnolam>
STOP! Foley time.
14:55 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
15:36
< Anno[Laptop]>
Hmm. If I wanted to run DF in high priority on one core only, would my command be: "nice -0 taskset -c 1 ./libs/Dwarf_Fortress"?
15:36
< Anno[Laptop]>
Or the other way around?
15:40
< Rhamphoryncus>
I'd to taskset first, just my preference though
15:41
< Rhamphoryncus>
*unless* taskset can only handle a single child process and is not inherited. Then it must be second
15:43
< Anno[Laptop]>
It appears to work the way I stated it.
15:44
< Anno[Laptop]>
Will this have a hardware effect, eg. overheating?
15:44
<@TheWatcher>
No
15:45
< Anno[Laptop]>
Cool.
15:45
< Anno[Laptop]>
DF is constantly running at ~100% CPU, according to top, though.
15:45
< Anno[Laptop]>
Even when it probably shouldn't be mulling things over so much.
15:47
< gnolam>
Well, it's pretty obvious that DF is horribly coded. :P
15:48
< gnolam>
It's probably busy waiting somewhere...
15:54
< Anno[Laptop]>
Either way, it doesn't prevent me from Alt-Tabbing to chat, so it's fine.
15:55 Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code
16:43
<@ToxicFrog>
Anno[Laptop]: you told the OS to dedicate a single core entirely to DF; DF has no sleeps or other calls that would make it unschedulable; what do you expect?
16:43
< Tarinaky>
What did Anno do?
16:46
< Anno[Laptop]>
Everything.
16:48 * ToxicFrog gets to the results section \o/
16:48
<@ToxicFrog>
Microblaze: 440 packets/second
16:49
<@ToxicFrog>
Microblaze with coprocessors: 3,990 packets/second
16:49
<@ToxicFrog>
Server: 207,268 packets/second
16:49
<@ToxicFrog>
RTL: 6,175,454 packets/second
16:55
<@ToxicFrog>
o.O
16:57
< Rhamphoryncus>
Anno[Laptop]: You may want to manually clock down that core
16:58
< Anno[Laptop]>
How?
16:58 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
16:58
< Rhamphoryncus>
If you're in gnome you can add a cpu frequency scaling monitor to your panel
16:59
< Rhamphoryncus>
although it only handles one cpu at a time, which is a bit silly. I created a drawer and have 4 of them
17:00
< Rhamphoryncus>
Right click on it and click preferences to set which CPU, then left click to change the frequency
17:00
< Anno[Laptop]>
How do I add the monitor to the panel?
17:01
< Rhamphoryncus>
right click on the panel, select add to panel, search for cpu
17:02
< Anno[Laptop]>
Left clicking the icon does nothing.
17:02
<@ToxicFrog>
Rhamphoryncus: how will that help?
17:03
<@ToxicFrog>
AIUI his goal is to apply maximum power to DF.
17:03
< Anno[Laptop]>
At the same time, I don't want my computer to explode.
17:03
< Rhamphoryncus>
Since DF apparently uses 100% CPU regardless this would let you slow it down when not needed without slowing down the rest of the computer
17:04
<@ToxicFrog>
If your computer is going to explode due to running the processor at 100%, (1) get a refund, because you have been given defective components and (2) how did you play DF before, since it uses up 100% regardless?
17:04
< celticminstrel>
DF?
17:04
< Anno[Laptop]>
I did not know it always uses 100.
17:04
<@ToxicFrog>
Dwarf Fortress.
17:04
<@ToxicFrog>
Most games do.
17:04
< Tarinaky>
Rhamphoryncus: Don't both cores have to be clocked at the same speed?
17:04
<@ToxicFrog>
DF is one of them.
17:04
< Rhamphoryncus>
You're left clicking the applet after adding it to the panel?
17:05
< Rhamphoryncus>
Tarinaky: no, maybe with some CPUs
17:05
< Anno[Laptop]>
Yes.
17:05
<@ToxicFrog>
This is why I was confused about why you even bothered using nice earlier - DF will use the entire core anyways.
17:05
< Rhamphoryncus>
weird
17:06
< Rhamphoryncus>
true, it's not quite sensible
17:07
<@ToxicFrog>
Anno[Laptop]: bear in mind that there's like four different things with "CPU monitor" in the name, make sure it's "CPU frequency scaling monitor"
17:07
< Anno[Laptop]>
I did.
17:09
< Rhamphoryncus>
What do the preferences show you?
17:09
< Rhamphoryncus>
And what gnome version?
17:09
< Anno[Laptop]>
Choice of CPU, and whether to display in % or Hz.
17:10
< Anno[Laptop]>
2.22.3.
17:10
<@ToxicFrog>
That said, most of this is pointless. You don't need nice unless your system is overloaded (as in, all cores at 100% so the OS needs to start giving stuff less time than it asks for), because DF will use as much CPU time as available. Using the underclocking tool will hurt performance, and if you need to do that to stop your system from breaking you have a defective computer.
17:10
<@ToxicFrog>
Taskset might be useful, if your kernel likes to move it between cores a lot and this is affecting performance, but that's unlikely.
17:14
< Rhamphoryncus>
I'm all for experimenting, but I'm with ToxicFrog. It's not likely to have real benefits
17:14
< Anno[Laptop]>
DF seems to have started to run faster after I've started to use taskset.
17:22 Attilla [Some.Dude@Nightstar-b125b36f.threembb.co.uk] has joined #code
17:22 mode/#code [+o Attilla] by Reiver
17:32 kwsn [kwsn@Nightstar-1def9d24.dyn.centurytel.net] has quit [[NS] Quit: BEEP BEEP IMMA JEEP]
17:32 cpux_ [chatzilla@Nightstar-c978de34.dyn.optonline.net] has joined #code
17:33 Netsplit *.net <-> *.split quits: cpux, @Attilla, @Kazriko, Anno[Laptop], @McMartin, @jerith
17:33 cpux_ is now known as cpux
17:36 Netsplit over, joins: McMartin
17:37 mode/#code [+o McMartin] by Reiver
17:39 Anno[Laptop] [annodomini@Nightstar-8ddde7e1.adsl.tpnet.pl] has joined #code
17:41 jerith [jerith@Nightstar-bf52129d.slipgate.za.net] has joined #code
17:41 mode/#code [+o jerith] by Reiver
17:42 Attilla [Some.Dude@Nightstar-b125b36f.threembb.co.uk] has joined #code
17:42 mode/#code [+o Attilla] by Reiver
18:33 Taki^ [jwjw@Nightstar-726bee87.consolidated.net] has quit [[NS] Quit: ]
18:44 Kazriko [kaz@Nightstar-e09690fa.client.bresnan.net] has joined #code
18:44 mode/#code [+o Kazriko] by Reiver
18:44 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
18:54 Attilla [Some.Dude@Nightstar-b125b36f.threembb.co.uk] has quit [[NS] Quit: ]
18:56 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
19:01 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
--- Log closed Mon Oct 25 19:25:07 2010
--- Log opened Mon Oct 25 19:25:28 2010
19:25 TheWatcher [chris@Nightstar-b4529b0c.zen.co.uk] has joined #code
19:25 Irssi: #code: Total of 24 nicks [5 ops, 0 halfops, 0 voices, 19 normal]
19:25 mode/#code [+o TheWatcher] by Reiver
19:26 Irssi: Join to #code was synced in 54 secs
19:31 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
19:36 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
19:47 Netsplit *.net <-> *.split quits: @Kazriko
19:49 Netsplit over, joins: Kazriko
19:49 mode/#code [+o Kazriko] by Reiver
20:00 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
20:04 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
20:26 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
20:52 Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has joined #code
20:52 mode/#code [+o Derakon] by Reiver
20:52 * Derakon mutters at OpenGL.
20:52
<@Derakon>
Okay, so I got an "invalid operation" with my code. How was it invalid?
20:52
<@Derakon>
The call conforms to the documentation in terms of arguments passed, so I'm guessing my context was wrong somehow?
20:55
<@Derakon>
(Unrelated: I love how in the required post-session status email users will say everything was fine, and then later they'll come up to me in person and ask about specific problems they encountered)
20:55 TarinakyKai [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code
20:58 Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Ping timeout: 121 seconds]
21:06
<@Derakon>
Ahh, there we go; I believe I now have a functioning render-to-framebuffer system.
21:21 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
21:25 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
21:31 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
21:35 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
21:39 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
21:40 SmithKurosaki [Smith@Nightstar-8ff23d84.dsl.teksavvy.com] has joined #code
21:43
<@Derakon>
Aaaaaaargh, this software is a CROCK OF SHIT
21:44 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
21:44
<@Derakon>
The idea of having a standalone package that has just about everything you need to do scientific computing with Python is good, but it needs someone remotely COMPETENT at the helm.
21:46
<@jerith>
Derakon: scipy?
21:46
<@Derakon>
Jerith: and wx, and FFTW, and plotting, and a bunch of other crap too.
21:47
<@Derakon>
And every single version I've found is different from all the other versions.
21:47
<@Derakon>
Well.
21:47
<@Derakon>
Except for the versions that I've installed over previous versions, which new ones were all copied from one original.
21:48
<@Derakon>
The problem here is that I have to have the damned thing or I can't read the data files the program needs (which are in a custom microscopy/astronomy format), but it hauls all this other crap along for the ride which means that I can't use any of it!
21:49
<@Derakon>
Its numpy install appears to be completely broken, for example.
21:50 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
21:51 * jerith sads.
21:55 Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
21:55
<@Derakon>
The main problem here is that my boss is breathing down my neck to get a copy to his buddies in Paris, and apparently had assumed that I'd been developing for Linux this entire time.
21:56 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
21:56
<@jerith>
:-(
21:56
<@Derakon>
Despite the fact that all of our users do all of their work and store all of their data on OSX computers -- so working on Linux would mean they'd start the program up over a remote X session, transfer all their data over network drives, render it to a virtual display, transfer the pixels back to their main computer, etc.
21:57
<@jerith>
Butbutbut! Mac OS X is Linux! (Isn't it?)
21:57
<@Derakon>
BSD, mate.
21:58
<@jerith>
Did you tell him that?
21:59
<@jerith>
(My work laptop is a Mac, and it's been my primary machine for a year and a half.)
22:00
<@Derakon>
He sent me an email saying, and I quote, "Its news to me that the edit program is written for Macs--very few of our users are mac users for data, macs are used for email etc."
22:00
<@jerith>
(I much prefer Linux, but switching between them is a recipe for never knowing what keychords to hit to do something.)
22:00
<@Derakon>
Which tells me he doesn't pay much attention to what users are doing day-to-day.
22:01
<@Derakon>
All of the data I've seen is stored on external hard drives connected to various flavors of OSX boxen.
22:02 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
22:02
<@Derakon>
I think boss is still thinking like we have people in our actual lab who use the scope, but our last biologist left almost a year ago. All of our users are from other labs now.
22:02
<@jerith>
Hrm.
22:03
<@jerith>
How much work is it to port?
22:03
<@jerith>
My initial reflex is "Python, so not much", but that's with deliberately limited GUI experience.
22:03
<@Derakon>
Theoretically, all you do is install the right libraries on your target platform and run the code.
22:03
<@Derakon>
Unfortunately we have the aforementioned CROCK OF SHIT.
22:04
<@Derakon>
Which is not cross-platform-compatible.
22:04
<@Derakon>
It's not anything-compatible.
22:05
<@jerith>
It would have to be if you're invoking Brian Blessed every time you mention it.
22:06
<@Derakon>
Apologies, I'm a bit aggravated.
22:07 * jerith isn't bothered.
22:07 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
22:08
<@jerith>
I just like the idea of a CS geek calmly explaining the problem with something and having Brian Blessed leap in front of him to yell its name every time it's mentioned.
22:09
<@Derakon>
Tone dissonance! It's what I do.
22:10 * jerith grins AT! the thing.
22:24
<@ToxicFrog>
Could be worse, you could have that and Visual Studio
22:25
<@Derakon>
Windows only shows up for the microscope systems.
22:25
<@Derakon>
And amazingly enough those seem to actually work most of the time.
22:28 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
22:33 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
22:41 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [Ping timeout: 121 seconds]
22:46 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
23:01 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [[NS] Quit: ]
23:06
< gnolam>
ToxicFrog: ?
23:07
<@ToxicFrog>
?
23:07
< gnolam>
What's wrong with Visual Studio?
23:09
<@ToxicFrog>
It has a godawful UI and huge configuration overhead for even the smallest of projects.
23:13
< gnolam>
It's a godsend for large projects.
23:13
< gnolam>
And for small, one-off programs, you're probably best off using another language anyway.
23:15 You're now known as TheWatcher[T-2]
23:18 You're now known as TheWatcher[zZzZ]
23:21 Attilla [Some.Dude@Nightstar-c253c6f1.threembb.co.uk] has joined #code
23:21 mode/#code [+o Attilla] by Reiver
23:25
<@ToxicFrog>
Interestingly, the only -really- large project I've worked on (CacheOS) moved away from Visual Studio several years ago, both because it was a support nightmare and because it was too inflexible to accomodate our requirements.
23:27
<@ToxicFrog>
That said, I can see how IDEs, in general, are great for medium-to-large projects.
23:27
<@ToxicFrog>
That doesn't change the fact that VS's interface is a crime against god and man.
23:27
< gnolam>
Eh. I just ignore all the parts I don't use.
23:28
< gnolam>
And even if the parts I used had been awful, which I don't find them to be, it would have been worth it all for the debugger.
23:29
<@ToxicFrog>
Regrettably, I actually have to to things like add third-party libraries and headers, which means I cannot ignore its hideous configuration clusterfuck.
23:30
<@ToxicFrog>
I do agree that the debugger is pretty handy, but it's not so great that it's worth the pain of the rest of it, for me.
23:32
<@ToxicFrog>
I mean, on the one hand I could use VS and get a nice debugger; on the other hand I could use anything else and get an adequate debugger, along with not having to worry about stub libraries or precompiled headers or which runtime version I'm linking against or wading through dozens of pages of what is possibly the worst configuration UI I have ever been subjected to.
23:32
<@ToxicFrog>
Seriously, stub libraries in this day and age, WTF
23:32
<@Derakon>
I just tend to use print statements. ?.?
23:33
<@ToxicFrog>
You are a bad programmer and should feel bad~
23:33
<@Derakon>
Those and a good stack trace tend to pin things down quickly.
23:47
<@ToxicFrog>
Yeah, but wouldn't it be better to pin things down quickly without needing the print statements in the first place?
23:51 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has joined #code
23:55 Taki^ [ikat@Nightstar-726bee87.consolidated.net] has quit [[NS] Quit: ]
23:59 Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has quit [[NS] Quit: Leaving]
--- Log closed Tue Oct 26 00:00:16 2010
code logs -> 2010 -> Mon, 25 Oct 2010< code.20101024.log - code.20101026.log >