code logs -> 2010 -> Fri, 29 Oct 2010< code.20101028.log - code.20101030.log >
--- Log opened Fri Oct 29 00:00:20 2010
00:25 You're now known as TheWatcher[T-2]
00:27 Attilla [Some.Dude@Nightstar-660d70eb.threembb.co.uk] has quit [[NS] Quit: ]
00:27 You're now known as TheWatcher[zZzZ]
00:38
<@ToxicFrog>
kazrikna: same build system. Different environment because I'm running it as me (so the dude who hardcoded paths leading into his home directory is SOL) and building from a clean checkout (so the dude whose testing presumably worked only because he had a binary lying around from earlier is likewise SOL)
00:40
<@ToxicFrog>
Both of these are things you can and should test yourself.
00:41
<@ToxicFrog>
And no, they didn't forget files, they just wrote build systems that were dependent on parts of their personal environment or home directory.
00:41
<@ToxicFrog>
...to clarify, parts that either they can't check in (like LD_LIBRARY_PATH) or parts that they shouldn't check in (like copies of system libraries or binaries generated from earlier builds)
00:43
<@ToxicFrog>
There's also the ones with makefiles that work, but report failure, because they used rm rather than rm -f to remove some file that is not generated in the normal course of usage.
00:46
<@kazrikna>
ahh, so some of them are stupid errors that they couldn't actually test because they only have 1 account to test with?
00:46
<@kazrikna>
Some are ones they could test if they made a second folder...
00:47
<@kazrikna>
The first would be a rather dumb thing, but I can imagine them thinking it's working perfectly because they only have their own folder.
00:50
<@ToxicFrog>
They only have one account, but they have done a test build on their home machines, one of the other clusters, or one of the lab machines.
00:51
<@ToxicFrog>
And if they haven't figured out "test it on machines other than the one on which you developed" and "test it with a clean checkout" by now, I just don't know.
00:51
<@ToxicFrog>
Also, I probably would not have penalized them as harshly for this shit if they had documented it.
00:52 * kazrikna wonders how many have the same username on their home machine...
00:52
<@ToxicFrog>
Your program needs a custom LD_LIBRARY_PATH and has a nonstandard build process? Ok, you lose 0.5pts for each of those. Oh wait, none of this is documented anywhere and I need to figure it out by trial and error and reverse-engineering your rat's nest of makefiles? It's on.
00:53
<@ToxicFrog>
kazrikna: the problem is not username, it's things like "I copied the system libraries into my home directory and then hard-coded the makefile to point at those"
00:53
<@kazrikna>
Eh. the rest of that... LD paths and such... Yeah, should be documented.
00:53
<@kazrikna>
Ahh. that would be a bit odd.
00:54
<@ToxicFrog>
Or "'make' at the top level doesn't actually build the program, 'make' or 'make wp' in src fails with a syntax error, 'make main' succeeds but generates a program in the wrong place with the wrong name"
00:54 * kazrikna works in a system right now that all of the system libraries are uniform and automatically included by the sample makefiles...
00:55
<@kazrikna>
Embedded stuff... usually more of an adventure than normal OS programming, but...
00:55
<@ToxicFrog>
Oh, we gave them a sample makefile with all of the necessary CCFLAGS and LDFLAGS and sample rules for building programs and libraries and whatnot.
00:55
<@ToxicFrog>
And it turns out that the students who made the minimal changes necessary to the sample makefile did not lose marks for having broken build systems!
00:55
<@kazrikna>
Heh.
00:56
<@ToxicFrog>
Except for the one who apparently erased 'make clean' from his and the three who didn't bother to alter the default provided, but the penalty for 'make clean doesn't work' is extremely minor whereas 'your program doesn't build' is potentially a 0.
00:57
<@kazrikna>
So, noting these down so you can give many more points down for repeats of the exact same problem? :)
00:59
<@ToxicFrog>
Yep. Although build problems are unlikely to recur simply by virtue of the fact that the next two assignments use a completely different environment.
01:00 Orthia [orthianz@Nightstar-2a4dd775.xnet.co.nz] has quit [Ping timeout: 121 seconds]
01:01
<@ToxicFrog>
That said.
01:02
<@ToxicFrog>
These are fourth year students operating in an environment very similar to the one they've been using since their first semester.
01:02
<@ToxicFrog>
They have no excuse whatsoever and I seriously considered awarding a 0 for any build issues.
01:03
<@kazrikna>
Heh.
01:03
<@kazrikna>
I think I took my first 4th-year class when I was in my 2nd year at college...
01:03
<@kazrikna>
Theory of Algo I think.
01:05
<@kazrikna>
At least you're letting them use unix based systems. I had one professor who basically would give you a 0 if you didn't use her framework that was 100% Visual C++ dependant... and I was using OS/2 at that point with GCC and the standard unix tools.
01:06
<@kazrikna>
I didn't even have a windows machine to work on...
01:07
<@Namegduf>
ToxicFrog: Let them you're awarding zero, then relent? :P
01:07
<@McMartin>
Namegduf: Sets a bad precedent that one can whine for grades.
01:07
<@ToxicFrog>
This is CIS*4450 Special Topics in Computing. You don't get in without talking personally to the prof and the prereqs this year mean you are at least in your third year.
01:08 Vornicus-Latens is now known as Vornicus
01:08
<@kazrikna>
I used to be a lab assistant in the computer lab though. I know some 4th year students that had a hard time putting together a valid for loop in C++
01:08
<@ToxicFrog>
;.;
01:08
<@kazrikna>
different college though. This one was fairly lenient in accepting students.
01:08
<@ToxicFrog>
But yeah, my point was more "no-one in this course is a raw newbie who somehow managed to sneak into an upper-year class"
01:09
<@kazrikna>
(which is why I was taking the 3rd and 4th year classes my second year.)
01:10
<@kazrikna>
I'm pretty sure at this point that my 7 years experience in the field is far more valuable on a resume than my college degree.
01:11
<@kazrikna>
Wish I would have stuck to the original plan and transferred after 2 years, but getting married changes things.
01:12 * ToxicFrog nods
01:16
<@ToxicFrog>
(environment-wise, engineering is mostly windows with some solaris, comp sci is mostly linux or OSX with some windows and BSD. Most of the work happens on the linux systems, which are remote-accessible.)
01:16
<@ToxicFrog>
(this course uses SHARCNET, which is not part of the labs but is linux-based)
01:17
<@kazrikna>
When I started there, they were a mix of Windows 3.11 and Windows 95, they mostly had windows 2000 by the time I got my job though.
01:18
<@kazrikna>
The only systems that were linux or unix were the ones in the professor's labs.
01:18
<@kazrikna>
and a couple of aging servers.
01:18
<@ToxicFrog>
o.O
01:18
<@ToxicFrog>
When was this?
01:19
<@kazrikna>
A few of us students built some linux systems and kept them in the csci lab.
01:19
<@kazrikna>
1997ish.
01:19 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
01:19
<@kazrikna>
Eesh. I guess I have 8 years experience at my job now. heh
01:20
<@kazrikna>
They hadn't exorcised all of the windows 3.11 stuff until at least late 1998.
01:21
<@kazrikna>
and they weren't all on windows 98 until at least 2000.
01:23
<@ToxicFrog>
;.;
01:23 Anno[Laptop] [annodomini@Nightstar-e49b2f7b.adsl.tpnet.pl] has quit [[NS] Quit: Way past my bedtime.]
01:23
<@kazrikna>
They only had a tiny handful of XP systems when I finished up in 2002.
01:24
<@kazrikna>
I spent a couple of summers upgrading 3.11 systems to 95 and 98 on campus. Scripts running and reading a book waiting for the ancient systems to catch up.
01:25
<@kazrikna>
State colleges never put enough money into their computers... heh
01:30 TarinakyKai [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code
01:32 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
01:34 * ToxicFrog ponders
01:34 Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Ping timeout: 121 seconds]
01:34
<@ToxicFrog>
Considering how badly this assignment went working in an environemtn they're familiar with, the next one is going to be an amazing clusterfuck
01:36
<@McMartin>
Maybe they'll do better because they won't be under the misapprehension that they already know everything?
01:36
<@Namegduf>
Any reason they'd be in an incredible rush or something?
01:39 TarinakyKai [Tarinaky@Nightstar-f349ca6d.plus.com] has quit [Connection closed]
01:44 Derakon[AFK] is now known as Derakon
01:53
<@Derakon>
Namegduf: they're students^Whuman. They'll always put it off to the last minute if they can.
01:53
<@Namegduf>
I mean especially so, though. Other deadlines and such.
01:55
<@Derakon>
It's my experience that with proper time management you almost never run into a situation where deadlines from other courses interfere with a specific course's work. It requires collusion across three departments at least.
01:55
<@Derakon>
This does require proper time management though.
01:55
<@Derakon>
I made it through college only having to stay up past midnight because of an assignment once.
01:56
<@Derakon>
This is practically freakish, as I understand it.
01:56
<@Namegduf>
I have staying up past midnight where I'm not forced not to as a default state, so it's a rough metric.
01:57
<@Namegduf>
*poor metric for me.
01:57
<@Namegduf>
With proper time management it shouldn't happen, yeah.
01:58
<@Namegduf>
I'm wondering if some bad management happened with a bunch of deadlines at all once, and the bonus course got a low priority.
01:58
<@Namegduf>
It seems unlikely this early in the year, though.
01:58
<@Derakon>
My point was that that would require a significantly large coincidence as to be unable to explain more than one or two students.
01:59
<@Namegduf>
Not if they're on similar sets of courses, which is the norm at some universities.
01:59
<@Namegduf>
There were a number of lectures last year which had about three people show up because of major deadlines in an hour or so.
01:59
<@Derakon>
Counting electives?
01:59
<@Namegduf>
Counting electives.
02:00
<@Derakon>
I mean, courses in the same department or even in related departments tended to work together to ensure reasonable lack of overlap in my college.
02:00
<@Namegduf>
Ah, I never had that.
02:00
<@Derakon>
I suppose it helps to have had a small college.
02:00
<@Namegduf>
They would quite happily schedule deadlines right after each other and leave it to the student to manage their time properly.
02:00
<@Namegduf>
Which either happened or it didn't, depending on who you were.
02:04
<@Derakon>
Hence teaching a valuable lesson in time management to some of the students. :)
02:04
<@Namegduf>
Yep.
02:07 Derakon is now known as Derakon[AFK]
02:17
<@ToxicFrog>
Namegduf: they have in fact received a week extension because they had two midterms the day before it was originally due.
02:28
<@Namegduf>
Ah.
02:29
<@kazrikna>
I still say, tell them what they messed up, and say they'll get double points off if they do it again.
02:32 Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has joined #code
02:33 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has quit [[NS] Quit: Z?]
02:36 Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has quit [Ping timeout: 121 seconds]
02:48 Derakon[AFK] is now known as Derakon
03:14 Zed [Zed@Nightstar-556ea8b5.or.comcast.net] has joined #code
04:00 Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has quit [Client exited]
04:45
<@ToxicFrog>
kazrikna: I won't actually be communicating the marks to them, but I provided a detailed breakdown of how I arrived at each mark to the prof.
04:50
<@kazrikna>
Oy.
04:50 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!]
04:51
<@ToxicFrog>
kazrikna: thing is, this is only half of the mark for the assignment; the other half is the report and analysis, which he's grading
04:56
<@kazrikna>
Looks like you're in for a fun semester of finding exactly the same problems over and over. :)
04:57
<@ToxicFrog>
I have confidence that he will effectively communicate my disappointment to them.
05:30
<@Derakon>
"Here's your scores for the most recent assignment. While some of you did well, I want you to know that some of you made the grader very sad."
05:41
<@McMartin>
"Every time you fail to read the spec and implement anyway, a kitten dies. Please. Think of the kittens."
05:42
<@kazrikna>
I think cats would go extinct within a decade if that were true.
05:44
<@Derakon>
Don't tell the Australians.
05:44 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Client closed the connection]
05:50 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
05:59 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Connection reset by peer]
06:07 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
06:08
<@kazrikna>
11:15 and I'm just about to go home from work...
06:08
<@McMartin>
Boo.
06:12
<@kazrikna>
39 hours by thursday, that means I'll probably get 16 hours overtime this week. :)
06:15
<@kazrikna>
Ok, home. :)
06:20 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Client closed the connection]
06:20 Derakon is now known as Derakon[AFK]
06:24 cpux is now known as shade_of_cpux
06:27 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
06:27 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code
06:27 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Connection reset by peer]
06:45 Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has joined #code
07:00 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
07:09 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Connection reset by peer]
07:17 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
07:17 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Client closed the connection]
07:42 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
07:42 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Connection reset by peer]
07:42 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has joined #code
07:53 McMartin [mcmartin@Nightstar-e77cc72c.pltn13.sbcglobal.net] has quit [[NS] Quit: network reset]
08:14 kwsn [kwsn@Nightstar-1def9d24.dyn.centurytel.net] has quit [[NS] Quit: BEEP BEEP IMMA JEEP]
08:26 McMartin [mcmartin@Nightstar-253cbfbb.pltn13.sbcglobal.net] has joined #code
08:26 mode/#code [+o McMartin] by Reiver
08:32 Vornicus is now known as Vornicus-Latens
08:34 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Ping timeout: 121 seconds]
08:37 gnolam [lenin@Nightstar-38637aa0.priv.bahnhof.se] has joined #code
08:42 Anno[Laptop] [annodomini@Nightstar-c5997e2c.adsl.tpnet.pl] has joined #code
08:51 Stalker [Z@3A600C.A966FF.5BF32D.8E7ABA] has quit [Ping timeout: 121 seconds]
09:10 You're now known as TheWatcher
09:32 Anno[Laptop] [annodomini@Nightstar-c5997e2c.adsl.tpnet.pl] has quit [[NS] Quit: leaving]
09:35 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code
10:34 Orthia [orthianz@Nightstar-f30997b2.xnet.co.nz] has quit [Ping timeout: 121 seconds]
11:20 Anno[Laptop] [annodomini@F67919.F326B3.98D923.BDA7B6] has joined #code
11:37 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has joined #code
12:18 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has quit [Client closed the connection]
12:26 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has joined #code
12:26 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has quit [Client closed the connection]
12:43 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has joined #code
12:44 Tarinaky [Tarinaky@Nightstar-f349ca6d.plus.com] has joined #code
12:51 Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has joined #code
13:06 Anno[Laptop] [annodomini@F67919.F326B3.98D923.BDA7B6] has quit [[NS] Quit: leaving]
13:20 Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has joined #code
13:43 Anno[Laptop] [annodomini@Nightstar-416fd987.adsl.tpnet.pl] has joined #code
14:13 celticminstrel [celticminstre@Nightstar-f8b608eb.cable.rogers.com] has joined #code
14:56 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!]
15:10 Orthia [orthianz@Nightstar-660be2ae.xnet.co.nz] has quit [Ping timeout: 121 seconds]
15:34 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has joined #code
15:56 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed]
16:49 Attilla [Some.Dude@Nightstar-b075dda3.threembb.co.uk] has joined #code
16:49 mode/#code [+o Attilla] by Reiver
17:48 Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has quit [Ping timeout: 121 seconds]
17:51 Attilla [Some.Dude@Nightstar-b075dda3.threembb.co.uk] has quit [[NS] Quit: ]
19:00 Rhamphoryncus [rhamph@Nightstar-473f8685.abhsia.telus.net] has quit [Client exited]
19:23 Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has joined #code
19:32 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has joined #code
20:36 Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has joined #code
20:36 mode/#code [+o Derakon] by Reiver
20:37
<@Derakon>
I love having commandline skills.
20:38
<@Derakon>
I just crafted a one-liner to help a coworker copy all of his C code files from throughout his workspace to a new directory, while preserving the directory structure.
20:38
<@Derakon>
(Granted, much of the work was done in Perl)
20:39
<@kazrikna>
heh.
20:39
<@Derakon>
find . -name "*.c" | perl -ne 'chomp; /(.*)\/(.*)/; `mkdir -p $1`; `cp $_ backups/$_`'
20:41
<@Derakon>
... http://geociti.es/
20:42
<@McMartin>
That has to be a malware site~
20:42
<@Derakon>
Actually it appears to be an attempt to mirror about 1TB worth of actual Geocities sites.
20:44
<@Vornicus-Latens>
...why
20:46
<@Derakon>
Achaeology, I'd guess.
20:47 Vornicus-Latens is now known as Vornicus
20:57 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has quit [Connection reset by peer]
21:05 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has joined #code
21:06 * Derakon mutters at his current project.
21:06 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has quit [Connection reset by peer]
21:07
<@Derakon>
I hate debugging OpenGL drawing code; it's so opaque.
21:13
<@Derakon>
Granted that what I'm trying to do is rather complicated.
21:13 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has joined #code
21:15
<@Derakon>
Specifically, given a large OpenGL canvas (e.g. 10000x10000 pixels) shrunk down to fit into a small viewing area, I want to pre-render large numbers of textures at that zoomed-out level.
21:15
<@Derakon>
http://paste.ubuntu.com/522278/ is my current code.
21:16
<@Derakon>
'scuse me, http://paste.ubuntu.com/522279/ is.
21:16
<@Derakon>
Which generates results that look like this: http://derakon.dyndns.org/~chriswei/temp2/framebuffer.png
21:17
<@Derakon>
(Ignore the bordering, which is an artifact of my screenshot)
21:17
<@Derakon>
Basically, the tiles visible along the right and bottom borders are not included in the pre-rendered version, and thus are rendered individually. They can be assumed to be rendered in the correct location.
21:17
<@Derakon>
All of the other tiles are in the pre-rendered bit, and are vertically offset.
21:17
<@Derakon>
By about 75% of the width of one tile.
21:17
<@Derakon>
I'd love to know why.
21:22
<@Derakon>
What I really don't understand about this offset is that the rendering to the framebuffer is using exactly the same parameters as the normal rendering, so where is the offset coming from?
21:23
<@kazrikna>
I'll have to take a look when I get off work.
21:24
<@TheWatcher>
Dera: silly question, but have you checked that you are getting the coords you expect by printing out the values you're sending to glVertex2f?
21:25
<@Derakon>
Er, no.
21:32
<@Derakon>
Okay, debugging output indicates that when I do the prerender, the vertical position of the bottom tile is at 16.5x the size of one tile; the vertical position of the non-prerendered tile below it is 18x the size of one tile. This is as it should be.
21:33
<@Derakon>
(The original grid is a 12x12 grid of 384x384 tiles rendered every 576 pixels, so there should be a gap between each tile of 192px)
21:33
<@Derakon>
Which implies that I'm somehow rendering the mega-texture wrong, but it's simply being rendered to cover the entire canvas.
21:37
<@Derakon>
Oooh, I've discovered an optical illusion. http://derakon.dyndns.org/~chriswei/temp2/optical.png
21:38
<@Namegduf>
My eyes are glitching.
21:44
<@jerith>
That makes my head hurt. :-(
21:45
<@Derakon>
What I find interesting is that without the borders around each gradient, the effect isn't there.
21:46 * Vornicus gets a slight "darken" effect as he moves his head to the right.
21:46
<@Vornicus>
but that's it.
21:47
<@Derakon>
I see the aisles between each tile as being wider to the left.
21:48
<@Derakon>
Alternately, the boxes look like trapezoids that are narrower on the left.
21:54
<@Vornicus>
huh. Yeah I don't see that one.
21:55
<@Derakon>
It goes away for me if I make the border any wider.
22:06 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has quit [Connection reset by peer]
22:13 Orthia [orthianz@Nightstar-e4d805d0.xnet.co.nz] has joined #code
22:28 * Derakon turns things over to the GameDev.net guys, as he is, for now, stumped.
22:43 You're now known as TheWatcher[T-2]
22:45 You're now known as TheWatcher[zZzZ]
23:16 Thaqui [Thaqui@27B34E.D54D49.F53FA1.6A113C] has quit [Connection closed]
23:28 Derakon [Derakon@Nightstar-1ffd02e6.ucsf.edu] has quit [[NS] Quit: Leaving]
23:54 Stalker [Z@26ECB6.A4B64C.298B52.D80DA0] has quit [Client closed the connection]
--- Log closed Sat Oct 30 00:00:21 2010
code logs -> 2010 -> Fri, 29 Oct 2010< code.20101028.log - code.20101030.log >