code logs -> 2016 -> Thu, 22 Sep 2016< code.20160921.log - code.20160923.log >
--- Log opened Thu Sep 22 00:00:21 2016
00:01
<@celticminstrel>
Vornotron: Any idea why the order tl-tr-bl tr-br-bl produces crosses but tl-tr-bl tl-br-bl doesn't?
00:02
< Vornotron>
misnaming?
00:02
<@celticminstrel>
Hmm...
00:03
<@celticminstrel>
Well, tl is {x1,y1,z1} and tr is {x2,y1,z1}...
00:03
<@celticminstrel>
And bl is {x2,y2,z3} and br is {x1,y2,z4}, and (x1,y1) is (x*quad_size,y*quad_size) while (x2,y2) is (x1+quad_size,y1+quad_size)...
00:04
< Vornotron>
ah ah ah
00:04
< Vornotron>
look at bl and br's X coordinates
00:05
<@celticminstrel>
Oh.
00:05
<@celticminstrel>
Because at some point I switched the order of definition of those two.
00:05
<@celticminstrel>
Okay, if I swap those names, it works. \o/
00:06
<@celticminstrel>
I have short folds apparently. Not too worried about it.
00:06
<@celticminstrel>
I can always change it later if I decide to.
00:18
<@celticminstrel>
These implementations are one-liners. I can't believe SFML thinks its simpler not to have them. :/
00:20 Kylo-Ren [cmnd@Nightstar-7i6ec2.dyn.optonline.net] has joined #code
00:27
<@celticminstrel>
So now I have normals, but I still can't see any contours. I guess first thing to do is check that the normals are correct...
00:29
<@celticminstrel>
I suppose the normal for a flat,level surface should be (0,0,1) rather than (0,0,484), but apart from that it looks okay...
00:31
<@celticminstrel>
Vector norm is x . x if I recall correctly...
00:33
<@celticminstrel>
Wait no, that's not right.
00:34
<@celticminstrel>
Square root of that.
00:35 Kindamoody is now known as Kindamoody[zZz]
00:36
<@celticminstrel>
Still no shading after normalizing the normals though...
00:39 catadroid [catadroid@Nightstar-ig7mi6.dab.02.net] has joined #code
00:39
<@celticminstrel>
Something is giving a GL error. Maybe that's the reason this isn't working.
00:43
<@celticminstrel>
glDrawArrays, apparently. :|
00:48
<@celticminstrel>
...it's an error that's not even defined in GL.h.
00:48
<@celticminstrel>
But basically it looks like it's too big.
00:58
<@celticminstrel>
I don't really get this. It seems to work fine, yet it sets an error flag, and drawing in batches doesn't change anything.
01:00
<@celticminstrel>
Negating all my normals makes everything darker. o.o
01:00
< Vornotron>
of course. negated normals are pointing the wrong way
01:01 Kylo-Ren [cmnd@Nightstar-7i6ec2.dyn.optonline.net] has quit [[NS] Quit: - nbs-irc 2.39 - www.nbs-irc.net -]
01:01
<@celticminstrel>
Well, I tried it in case I already somehow had them pointing the wrong way, but I guess that's a sign that I didn't.
01:02
< Vornotron>
that shouldn't be an error source anyway.
01:03
<@celticminstrel>
Some of the normals do have -0, but I'd expec that to be irrelevant.
01:03
< Vornotron>
also irrelevant, yes
01:04
<@celticminstrel>
They're now primarily (0,0,1)... *skipping through the debugger*
01:04
< Vornotron>
that's what they should be for a predominantly flat thing
01:05
<@celticminstrel>
Here's one that's (-0.83205, 0, 0.5547).
01:06
< Vornotron>
slope is -1.5
01:07
<@celticminstrel>
I also see a (-0.639602, 0.639602, 0.426401). Nothing seems out of place to me...
01:08
<@celticminstrel>
That one's a quad with the top right corner raised.
01:08
<@celticminstrel>
Well, I'm calling it "top right", not sure if that's how it appears in the actual rendering.
01:09
<@celticminstrel>
(I seem to recall that "northwest" seems to be straight up and "northeast" to the left, so I probably need to change my coordinate systems or something.)
01:10
<@celticminstrel>
Not gonna step through every single vertex, since they number in the thousands.
01:13
<@celticminstrel>
Commenting out the light position gives black artifacts?
01:14
<@celticminstrel>
I'm actually wondering if I should use a sphere with emission rather than glLight.
01:19
< Vornotron>
that sounds like it wouldn't work at all?
01:20
<@celticminstrel>
It does? Okay then.
01:20
< Vornotron>
like, opengl doesn't do raycasting typically
01:20
<@celticminstrel>
Okay.
01:22
<@celticminstrel>
Aha! I need to have non-zero diffuse.
01:23
< Vornotron>
...yes
01:23
<@celticminstrel>
Okay now I just need to figure out good parameters, I guess.
01:24
< Vornotron>
I can't remember all the ones but there's ambient (dependent on no lights at all), diffuse (dependent on incoming light normal vs surface), and then specular & phong which are subtle enough difference that I don't remember what they are, but they depend on camera direction as well
01:25
<@celticminstrel>
I think phong isn't built-in to OpenGL...
01:25
<@celticminstrel>
Specular just seems to make the whole scene bright.
01:26
<@celticminstrel>
From what I've been reading it seems like it's supposed to have something to do with reflectivity.
01:27
<@celticminstrel>
Not sure what it means when applied to the light source rather than the material, though.
01:28
<@celticminstrel>
Hills still blend with the flat ground beyond them, but this is progress.
01:29
<@celticminstrel>
beyond -> behind
01:31
< Vornotron>
for that you may try: keep a bit of wireframe, to give reference-object perspective; mild depth fog; edge detection like borderlands..
01:32
<@celticminstrel>
Not sure what you mean by the last one.
01:32
<@celticminstrel>
I'd imagine a wireframe would help, yes.
01:32
<@celticminstrel>
I'm not sure if the problem would go away once textures are applied.
01:33
<@celticminstrel>
Didn't think of fog.,,
01:34
< Vornotron>
borderlands has this system
01:34
<@celticminstrel>
Normalizing 0..255 to floats - divide by 255 or by 256?
01:34
< Vornotron>
256
01:35
< Vornotron>
https://images-na.ssl-images-amazon.com/images/G/01/videogames/detail-page/borde rlands.01.lg.jpg it's clearest around Brick there.
01:37
<@celticminstrel>
Not sure what I'm looking for.
01:38
< Vornotron>
see the black edge around people?
01:38
< Vornotron>
especially the character on the left?
01:38
<@celticminstrel>
Ah. Yeah.
01:39
< Vornotron>
It's done, iirc, with a shader, checking nearby depth values
01:41
< Vornotron>
though I saw a thing once where they described how you could do it pre-shaders. Loooong time ago
01:55 * ToxicFrog looks at his code
01:56
<&ToxicFrog>
How does this possibly work?
01:56
<&ToxicFrog>
Ever?
01:57
<&ToxicFrog>
Like, here's the only code in the whole thing that sets the 'killed' field:
01:57
<&ToxicFrog>
match = re.match(r' was .* by (a|an|the) (.*) (on level|at the)', line)
01:57
<&ToxicFrog>
if match:
01:57
<&ToxicFrog>
mortem['killed'] = match.group(2)
01:58
<&ToxicFrog>
And yet somehow, a mortem containing " nuked the Mastermind at the Hell Fortress." ends up with mortem['killed'] == "nuked the Mastermind"
01:58
< Vornotron>
:(
01:59
<&ToxicFrog>
And I cannot figure out how
02:00
< Vornotron>
what's the full line
02:00
<&ToxicFrog>
That is the full line.
02:00
< Vornotron>
what.
02:00
<&ToxicFrog>
Here's an example: http://phobos.ancilla.ca/players/Roonespism/1.mortem
02:01 gizmore [kvirc@Nightstar-0jicpq.dip0.t-ipconnect.de] has joined #code
02:01
<&ToxicFrog>
" completed 100 levels of torture on level 100 of Hell." also properly shows up as "completed 100 levels"
02:01
<&ToxicFrog>
(http://phobos.ancilla.ca/players/ToxicFrog/163.mortem)
02:29 catadroid` [catadroid@Nightstar-r2ptdr.dab.02.net] has joined #code
02:33 catadroid [catadroid@Nightstar-ig7mi6.dab.02.net] has quit [Ping timeout: 121 seconds]
02:46
<&ToxicFrog>
Ahahahahahahaha IT'S ALIIIIIIIIVE
03:41
<&ToxicFrog>
wow
03:41
<&Derakon>
Za?
03:41
<&ToxicFrog>
I have been working on getting this thing centered on the page for an hour now
03:41
<&ToxicFrog>
CSS delenda est
03:42
<&Derakon>
Ah yes, centering woes.
03:42
<&Derakon>
A.k.a. why I'm glad I'm not a web dev any more.
04:26
<@celticminstrel>
I wonder if there's any way to obtain a list of global symbols in my program...
04:26
<@celticminstrel>
Variables, not types.
04:26
<@celticminstrel>
Wait, XCode has a list?
05:00 gizmore [kvirc@Nightstar-0jicpq.dip0.t-ipconnect.de] has quit [Ping timeout: 121 seconds]
05:00
<&ToxicFrog>
Derakon: could you do me a quick favour?
05:00
<&Derakon>
Where's the server again?
05:00 gizmore [kvirc@Nightstar-3bt752.dip0.t-ipconnect.de] has joined #code
05:00
<&ToxicFrog>
http://phobos.ancilla.ca/tty.html
05:00
<&Derakon>
...oh, you sly bastard, you built a tty server into the browser.
05:00
<&ToxicFrog>
Tell me if that gives you the doomrl-server banner and guest prompt
05:00
<&ToxicFrog>
Yeah, that's what I've been working on the past few days
05:00
<&ToxicFrog>
Hence my CSS woes
05:00
<&Derakon>
It does, and also warns me that my terminal size is 0x0, which is smaller than recommended.
05:01
<&ToxicFrog>
Yeah, that's expected
05:01
<&ToxicFrog>
xterm in the browser doesn't properly report terminal size
05:01
<&ToxicFrog>
It will in fact work correctly if you 'replay'. I haven't tried actually playing a game in it yet, I just got it working.
05:03
<&Derakon>
Well, I have a little time before bed. Let's kick the tires.
05:03
<&ToxicFrog>
Thanks!
05:04
<&ToxicFrog>
I'm on my way to bed; I mostly just wanted to see if the page loaded correctly for people outside my LAN. But if you want to take it for a spin, please do :)
05:04
<&ToxicFrog>
(Note that this thing is a gruesome Rube Goldberg machine cobbled together from ancient machines I do not understand and can barely control. It is not just possible but likely that it will crap out in some exciting and unforseen way and eat your character.)
05:06
<&Derakon>
First notable thing is that the Escape key doesn't register reliably.
05:07
<&Derakon>
Or maybe it's just that I have to hit Esc twice for it to count.
05:07
<&Derakon>
Yeah, it's the latter.
05:17
<&Derakon>
Whelp, ded.
05:17
<&Derakon>
I am rusty as fuck.
05:17
<&Derakon>
Should probably have started with SoB instead of Reloader.
05:37 Derakon is now known as Derakon[AFK]
05:47 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has quit [Connection closed]
05:47 VirusJTG [VirusJTG@Nightstar-6i5vf7.sta.comporium.net] has joined #code
05:47 mode/#code [+ao VirusJTG VirusJTG] by ChanServ
06:10 Kindamoody[zZz] is now known as Kindamoody
06:13 Vornotron [Vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
06:53 celticminstrel [celticminst@Nightstar-f4fn3d.dsl.bell.ca] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
07:18 Vornotron [Vorn@ServerAdministrator.Nightstar.Net] has joined #code
07:20 catadroid` [catadroid@Nightstar-r2ptdr.dab.02.net] has quit [[NS] Quit: Bye]
07:24 Vornotron [Vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
07:41 Kindamoody is now known as Kindamoody|afk
08:23 macdjord|wurk is now known as macdjord|slep
--- Log opened Thu Sep 22 09:52:43 2016
09:52 TheWatcher [chris@GlobalOperator.Nightstar.Net] has joined #code
09:52 Irssi: #code: Total of 37 nicks [32 ops, 0 halfops, 0 voices, 5 normal]
09:52 mode/#code [+o TheWatcher] by ChanServ
09:52 Reiver [quassel@Nightstar-ksqup0.co.uk] has joined #code
09:52 mode/#code [+ao Reiver Reiver] by ChanServ
09:53 You're now known as TheWatcher[lab]
09:53 Irssi: Join to #code was synced in 54 secs
10:42 catadroid [catadroid@Nightstar-e0ar9i.dab.02.net] has joined #code
10:42
< catadroid>
Python does not let me naturally format my code
10:42
< catadroid>
I wish to end it
10:42
< catadroid>
That is all
11:33
<@abudhabi>
"naturally format"?
11:35
< catadroid>
The fact that whitespace is significant makes it very hard to concatenate text format templates together in the way I would consider natural, since line breaks require more syntax
11:44
<&[R]>
Ugh
11:44
<&[R]>
Don't use templates to generate code please
11:44
<&[R]>
D:
11:45 * [R] has flashbacks from $CCP
11:54 You're now known as TheWatcher
11:57
<&ToxicFrog>
Derakon[AFK]: the escape key bug is a bug in doomrl itself and not in the terminal emulator.
12:22
< catadroid>
Templates to generate code? I'm using concatenating strings to output into files
12:22
< catadroid>
I'm concatenating*
12:24
<&[R]>
"concatenate text format templates"
12:24
<&[R]>
Eitherway, I'm pretty sure there has to be another way?
12:32 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
12:35
< catadroid>
I could spend days learning an xml writing framework or I could just write the text I know I need out
12:36
< catadroid>
In that case, building the text before outputting it is handy
12:37
< catadroid>
I would like to spread the lines of xml over multiple lines without duplicating syntax
12:37
< catadroid>
Syntactically significant whitespace is just awful :<
12:46
<@abudhabi>
Hrm.
12:47
<@abudhabi>
MySQL. I'm getting a list of numbers from a table A. I have table B, where there are numbers also, and I want to fetch some associated data.
12:47
<@abudhabi>
What I want is to get a list of numbers from table A, with the first associated hit on a field from table B.
12:48
<@abudhabi>
So if I've got 1234 from A, and I have [herp, derp] in B, then what I want to get is [1234, herp].
12:48
<@abudhabi>
I do NOT want to get [[1234,herp],[1234,derp]] which a join would get me.
12:49
<@abudhabi>
Hm. A subselect maybe.
12:49
<@abudhabi>
Select number, whatever from B where number is in A.
12:52 Emmy-werk [NSkiwiirc@Nightstar-41pbej.static.chello.nl] has joined #code
13:07 Vornicus [Vorn@ServerAdministrator.Nightstar.Net] has joined #code
13:07 mode/#code [+qo Vornicus Vornicus] by ChanServ
13:11
<&[R]>
Yeah, you'll either need a sub-select or multipe queries
13:34 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code
13:34 mode/#code [+ao ToxicFrog ToxicFrog] by ChanServ
13:37
< Emmy-werk>
Hmmmmh. Somewhat worried my current headache and ensuing difficulty concentrating will introduce bugs.
13:40
<@TheWatcher>
Do all the work on a new git "headache" branch, easier to isolate all the changes then~
13:41
< Emmy-werk>
heh.
13:42
< Emmy-werk>
would be a fun idea. that said, i still need to implement git some time. :/
13:43
< Emmy-werk>
Oh well. i quite methodically document changes i made, it'll be alright i think.
13:51
<@TheWatcher>
Eeeeh, "implement git" can just be a few minutes work, but
13:51
<@TheWatcher>
meh
13:52
< Emmy-werk>
indeed.
13:52
< Emmy-werk>
at the current scale of the project, and the amount of people working on it, (1: me) it's not that critical
13:52
<@TheWatcher>
I'd very strongly argue that, myself
13:53
<@TheWatcher>
There's a bunch of projects I work on, as the sole dev, from really simple single-script things up to 100,000 line monsters, and I wouldn't be without git for any of them. Saved my arse multiple times
13:56
< Emmy-werk>
Eh. It does automate a lot of things. at the moment i do these things by hand. does have its' advantages as well.
13:58
< catadroid>
I should learn how to git
13:58
< catadroid>
I just use subversion generally
13:58
<@TheWatcher>
I'm sorry~
13:59
< Emmy-werk>
no, your mom
14:03
<@TheWatcher>
(It's quite simple really, you just have to get used to thinking about the state of your repository as a point in high-dimensional code-space in which branches are represented as n-dimensional membranes mapping the spatial loci of successive commits onto the projected manifold of each cloned repository!)
14:06
<@gnolam>
A simple subset of analytical algebraic topology of local Euclidean metrization of infinitely differential Riemannian manifolds.
14:12
<@TheWatcher>
Only for Lobachevsky's version of git~
14:48 catadroid` [catadroid@Nightstar-g84906.dab.02.net] has joined #code
14:49
<&ToxicFrog>
catadroid: you shooooooould
14:49 * ToxicFrog snuggles catadroid
14:52 catadroid [catadroid@Nightstar-e0ar9i.dab.02.net] has quit [Ping timeout: 121 seconds]
14:53
<&ToxicFrog>
Emmy-werk: re git for single-person projects: I strongly recommend it; even for small things it's a convenient safety net.
15:34 macdjord|slep is now known as macdjord|wurk
15:55 mac [macdjord@Nightstar-tol.t6i.225.24.IP] has joined #code
15:55 mode/#code [+o mac] by ChanServ
15:58 macdjord|wurk [macdjord@Nightstar-r9vt2h.mc.videotron.ca] has quit [Ping timeout: 121 seconds]
15:59 Emmy-werk [NSkiwiirc@Nightstar-41pbej.static.chello.nl] has quit [[NS] Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
16:31 catadroid` is now known as catadroid
18:22 celticminstrel [celticminst@Nightstar-f4fn3d.dsl.bell.ca] has joined #code
18:22 mode/#code [+o celticminstrel] by ChanServ
19:08
<&McMartin>
The thing is, given your latest obsessions, the non-joke version of Watcher's line might actually help
19:08
<&McMartin>
A git repository is an evolving purely-functional graph
19:09
<&McMartin>
With each commit name as a reference to some version of it, unmodified by later changes. Anything that requires you alter those nodes instead creates new independent commits representing them, etc
19:09
<&McMartin>
That also said, for small single-person projects, git is being used as an on-site backup system, not as a DVCS
19:10
<&McMartin>
It has not escaped my notice that in the vast majority of cases, git scales to medium-sized organization scale largely by clumsily reinventing subversion~
19:14
<&ToxicFrog>
I would argue that even when you're reinventing svn with it, git is better at being svn than svn is
19:14 Pink [user1@Nightstar-g7hdo5.dyn.optonline.net] has quit [Connection reset by peer]
19:14
<&ToxicFrog>
And also that even for single person use, being a VCS implies different things than being a backup system, like easy patching/diffing
19:16
<&McMartin>
Yeah, I'm familiar with this argument
19:17
<&McMartin>
git is still so hopeless for blobs that if your system requires blobs the best way to deal with it is to maintain a separate svn repo just for blobs
19:18
<&[R]>
Blobs are horrible anyways.
19:18
<&McMartin>
Depending on what the blobs are you may be able to get away with a carefully structured, non-version-controlled, mountable disk partition that is occasionally added to but never has files removed or modified
19:18
<&McMartin>
R: Right, getting to say that drastically restricts the space of problems you are solving
19:18
<&McMartin>
Like, if you're making a commercial game program, you're out of luck
19:19
<&[R]>
Why are you versioning art assets though?
19:19
<&McMartin>
Saying "Well, assets shouldn't be synced to code revisions in the first place" is just insisting that this fatal problem isn't a problem, and there is a sneaking suspicion in the listener that this isn't a problem because git doesn't solve it
19:19
<@gnolam>
McMartin: thank you.
19:19
<&McMartin>
... for the same reason you version anything.
19:20
<&McMartin>
Art assets is in fact one of the cases where the non-version-controlled super disk doesn't work, in fact.
19:20
<&McMartin>
(That one works for "Every possible version of a precompiled third-party library you have ever relied upon" for languages that aren't Java)
19:20 Derakon[AFK] is now known as Derakon
19:20
<&[R]>
Okay, that's fair
19:21 Pink [user1@Nightstar-g7hdo5.dyn.optonline.net] has joined #code
19:22 * [R] hasn't had to deal with a situation where binary data had to be shipped with code. Has had situations where SQL would've been nice to include in the repo...
19:22
<&McMartin>
(I say "that aren't Java" because if you are using Java, Maven solves this problem better than manually provisioning a squintillion USB drives to everyone)
19:44
<@celticminstrel>
Maven seemed painful to use when I looked at it.
20:00 gizmore [kvirc@Nightstar-3bt752.dip0.t-ipconnect.de] has quit [Ping timeout: 121 seconds]
20:01 gizmore [kvirc@Nightstar-ofjk5v.dip0.t-ipconnect.de] has joined #code
20:26 catalyst [catalyst@Nightstar-bt5k4h.81.in-addr.arpa] has joined #code
20:27
<&McMartin>
Maven is incredibly painful to use
20:27
<&McMartin>
However, if you get it properly integrated into a development system, it is an invisible package manager that magically makes exactly what whatever it is you are building's dependencies mysteriously appear as needed and only ever downloads them once.
20:27
<&McMartin>
I freely admit that 'properly integrated' is doing a lot of heavy lifting in that sentence
20:28
<&McMartin>
But setting up a project to use maven transparently is nevertheless less work than maintaining a master list of versioned libraries independent of your project and updating it by hand.
20:42 Kindamoody|afk is now known as Kindamoody
21:03 Derakon_ [chriswei@Nightstar-5mvs4e.ca.comcast.net] has joined #code
21:05 Derakon [chriswei@Nightstar-5mvs4e.ca.comcast.net] has quit [Ping timeout: 121 seconds]
21:18 catadroid` [catadroid@Nightstar-eopd2k.dab.02.net] has joined #code
21:21 catadroid [catadroid@Nightstar-g84906.dab.02.net] has quit [Ping timeout: 121 seconds]
22:04 Kindamoody is now known as Kindamoody[zZz]
22:20 Alek [Alek@Nightstar-cltq0r.il.comcast.net] has quit [[NS] Quit: bbl]
22:33 ion [Owner@Nightstar-6grqph.vs.shawcable.net] has quit [Ping timeout: 121 seconds]
22:35 Alek [Alek@Nightstar-cltq0r.il.comcast.net] has joined #code
22:35 mode/#code [+o Alek] by ChanServ
22:35 Derakon_ is now known as Deraekon
22:35 Deraekon is now known as Derakon
22:35 mode/#code [+ao Derakon Derakon] by ChanServ
23:39 ion [Owner@Nightstar-6grqph.vs.shawcable.net] has joined #code
23:39 mode/#code [+o ion] by ChanServ
--- Log closed Fri Sep 23 00:00:44 2016
code logs -> 2016 -> Thu, 22 Sep 2016< code.20160921.log - code.20160923.log >

[ Latest log file ]