code logs -> 2012 -> Sat, 18 Feb 2012< code.20120217.log - code.20120219.log >
--- Log opened Sat Feb 18 00:00:21 2012
00:08
< maoranma>
It's amazing what you can find via HTTP with a little google magic
00:08
< maoranma>
allintitle: index of roms
00:23 iospace [Alexandria@Nightstar-635d16fc.org] has quit [Ping timeout: 121 seconds]
00:25 You're now known as TheWatcher[T-2]
00:30 You're now known as TheWatcher[zZzZ]
00:30 iospace [Alexandria@Nightstar-635d16fc.org] has joined #code
00:32 Derakon[AFK] is now known as Derakon
00:33
<&Derakon>
Finally got in an HDMI cable so I can switch my primary display from VGA to HDMI and use the spare VGA cable for a third display...except that my HMDI-capable display doesn't get any input signal when plugged in via HDMI cable. Dammit.
00:39
< maoranma>
HDCP?
00:39 * McMartin flays Windows Event Viewer, which has the slowest sorting time of anything he's ever seen
00:40
< maoranma>
That's cause it uses the while list: if unsorted: randomsort(list) technique
00:40
<&McMartin>
I'm not convinced it isn't
00:41
< maoranma>
it uses minimal code at least
00:41
<&McMartin>
Disagree; bubblesort is pretty much the absolute minimal
00:41
<&McMartin>
You need a PRNG to do bogosort
00:41
< maoranma>
How's that one work?
00:41
< maoranma>
That's true
00:42
<@Alek>
P?
00:42
<&McMartin>
Repeatedly scan through the list, swapping adjacent entries if they're out of order
00:42
<&McMartin>
Alek: Pseudo
00:42
<@Alek>
bah
00:42
<&McMartin>
Bubble sort is O(n) if it's sorted, and O(n^2) in all reasonable cases
00:42
< maoranma>
Hah
00:43
<&McMartin>
This does mean that it outperforms quicksort on a sorted list~
00:43
< maoranma>
I still like quantum suicide sort.
00:43
<&Derakon>
Bogosort is O(infinity)~
00:43
< maoranma>
In theory, since it could just run forever
00:44
<&McMartin>
Yup
00:49
< maoranma>
Oh, that gave me an idea
00:51
< maoranma>
I should write a python program that does bogosort (to max x iterations), bozosort, maybe goro sort, and "quantum bogosort" at the end
00:52
< maoranma>
I was wrting a dice module for extra credit, but I wonder which I should do now
01:01
<~Vornicus>
what is gorosort?
01:02
<~Vornicus>
oh, I see.
01:11
< Rhamphoryncus>
maoranma: add a pure python reimplementation of the timsort, but with a subtle bug to make it slower than the others for the inappropriately slow test cases you use :)
01:18
< maoranma>
Haha. I figured what I'd do with qbogosort is have it output the number of other universes destroyed ((2^N)-1)
01:20
< maoranma>
Hmm... which doesn't sound right...
01:23
< maoranma>
Let's see, it should spawn 2^n universes where in is the number of random bits. But at least of one should aways be sorted, so that's 2^0, and a list of two should be 2^1?
01:23 * maoranma goes back to his dice script...
01:25 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
01:26 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has joined #code
01:26 mode/#code [+qo Vornicus Vornicus] by ChanServ
01:28
< maoranma>
Here's a weird question, can I extrapolate info for a class based on the name of the object it's given... Ie, I have a class of dice, and say I do gurps = 3d6(), can I have the class figure out that by it's name it should roll 3d6 for itself?
01:29
<~Vornicus>
Good god don't fucking do that
01:29
< maoranma>
LMAO
01:30
< maoranma>
Actually, it would have to be 3d6 = diceObj()
01:30
<~Vornicus>
That won't work either anyway
01:30
< maoranma>
Then from self it can index out that info?
01:30
<~Vornicus>
Because 3d6 isn't a valid literal
01:30
< maoranma>
Oh, I see
01:31
<~Vornicus>
And no language that I am aware of tells a function or constructor what the name of the lvalue is.
01:32
< maoranma>
So, self in Python only return the name of the object, ie, the class, not the variable that's referencing it?
01:32
<~Vornicus>
It doesn't return a name at all.
01:32
<~Vornicus>
It's a reference; references aren't names.
01:32
< maoranma>
Ah, right
01:33
< maoranma>
So it's just a memory reference?
01:33
<&Derakon>
There could be any number of names for that object.
01:33
<&Derakon>
foo = Dice(); bar = foo; baz = bar;
01:33
< maoranma>
Yea, that's right
01:34
<&Derakon>
But more relevantly, if it were possible and you did this I'd have to kill you.
01:34
< maoranma>
Oh well, the concept question was worth it for Vornicus' reaction
01:35
< maoranma>
Always looking for ways to make my code as obtuse as possible
01:35
<~Vornicus>
Seriously why would you do that to yourself
01:35
< maoranma>
I'm automasochistic?
01:42
< maoranma>
Okay, so I want to define a class that has 2 nonoptional arguements and three entirely optional arguements, what would be the easiest way to handle that?
01:42
< maoranma>
Keyword arguements?
01:42
<~Vornicus>
sort(seq, reverse = False, key = None, cmp = None)
01:42
<&Derakon>
def __init__(self, arg1, arg2, arg3 = foo, arg4 = bar, arg5 = baz)
01:43
<~Vornicus>
Just like that
01:43
<&Derakon>
Just give defaults to all the optional parameters.
01:45
< maoranma>
so, to creat the object with say, just arg5, I would have to do obj(arg1,arg2,arg5=xyz)?
01:45
<&Derakon>
Uh, no...
01:45
<&Derakon>
In my example, that would throw an error "missing required parameter arg1".
01:46
<&Derakon>
Oh, wait, yes, that's what you'd do.
01:46
<&Derakon>
Sorry, misread.
01:46
< maoranma>
lol
01:46
<&Derakon>
It's Friday.
01:47
<&McMartin>
Three day weekend wooo
01:49
<&Derakon>
Yus.
01:49
<&Derakon>
And it's gonna be a busy one for me.
01:51
< maoranma>
So, if I did it positional like... def __init__(self,quantity,sides,exp=false,threshold=0,add="+0")... I should be able to create the obj = obj(3,6,0,true,"+1") with the last three in any order and it not melt?
01:51
< maoranma>
Since I have a bool, int, and string?
01:53
<&Derakon>
Uh, no.
01:53
<&Derakon>
Python doesn't have any type safety.
01:54
<&Derakon>
You have to use the order used in the function definition, or keyword your arguments.
01:54
< maoranma>
Ah
01:54
<&Derakon>
E.g. "obj(3, 6, threshold = 0, exp = true, add = "+1")"
01:55 Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has quit [Client exited]
02:00
<~Vornicus>
You can also, at that, keyword only a few options.
02:00
<~Vornicus>
obj(3,6,threshold = 4)
02:01
< maoranma>
Right
02:01
<~Vornicus>
Or non-keyword options, in order, but leave off the last few: obj(3,6,true,3)
02:03
< maoranma>
In order of common use, I'll put it add, threshold, exploding
02:03
<&Derakon>
Oh, is that what exp means?
02:03
< maoranma>
Exploding dice are something I've only seen a few times, but I like the concept of it
02:03
<&Derakon>
I assumed it meant exponent.
02:03
< maoranma>
nah, this is for my dice script
02:03
<~Vornicus>
by the way don't abbreviate your words unless they're very common abbreviations.
02:04
< maoranma>
Agreed
02:06
<&McMartin>
I think "exploding" for dice might be >_>
02:06
< maoranma>
Know what exploding dice are?
02:07
<~Vornicus>
Also, you might be violating a normal form here; is the threshold there the threshold to have the dice explode?
02:07 Derakon is now known as Derakon[AFK]
02:08
<&McMartin>
maoranma: Rolling an N on the N-sided die means you keep the N and add it to a reroll of the (also exploding) die?
02:09
< maoranma>
No, it's for rerolling, ie dice(3,6,threshold=1) means roll 3 6-sided dice, and reroll 1s
02:09
<~Vornicus>
Ah, call it reroll_threshold then.
02:09
< maoranma>
And yes, exploding dice is like that
02:09 Kindamoody[zZz] is now known as Kindamoody
02:10
< maoranma>
You could theoretically have exploding be a value instead of bool, so you can have it reroll and add any particular number, or range of numbers
02:11
< maoranma>
But the most common definition of it is reroll and add the highest side
02:11
< maoranma>
Vornicus: will do
02:15
< maoranma>
Oh, I meant to ask, is there a way to make a class perform it's methods on contruct?
02:25
<@ToxicFrog>
Call them from the constructor.
02:27
<~Vornicus>
Yeah. As long as the data that the method needs exists, you can use them all you want.
02:27
<&McMartin>
In Python, do not do this until after you have called super.__init__
02:28
< maoranma>
super.__init__?
02:28
<&McMartin>
The __init__ methods of your parent classes.
02:28
<&McMartin>
named super here for convenience
02:29
<&McMartin>
Whoops, d'oh
02:30 * McMartin has a working parser here but he forgot some fields in his AST, and as such needs to recode Many Things.
02:31
<~Vornicus>
note that if your parent is object, you don't have to worry about super
02:31
< maoranma>
I guess I'm not understanding. I get a global name 'x' is not defined whenever I do it where x is a function in the same class
02:32
<&McMartin>
You have to say self.x (assuming you've named the self argument self)
02:32
< maoranma>
Oh
02:32
< maoranma>
OH
02:32
< maoranma>
*headdesk* Of course it was that simple
02:33
< maoranma>
That self arguement throws me off every time
02:35
<&McMartin>
To be fair, this is a major difference between the Python family of OO (Smalltalk, Objective-C) and the more common one (C++, Java, C#).
02:49
< maoranma>
Oh, polymorphism, right
02:49
< maoranma>
x.append(y) != x + y
03:05 mode/#code [+o iospace] by ChanServ
03:24
< maoranma>
beep
03:28 * Vornicus beeps maoranma.
03:28 Kindamoody is now known as Kindamoody|gaming
03:31 Attilla [Obsolete@Nightstar-514682b9.as43234.net] has quit [Ping timeout: 121 seconds]
05:12 eckse [eckse@Nightstar-23b76b03.dsl.sentex.ca] has joined #code
05:12 mode/#code [+o eckse] by ChanServ
06:29 eckse [eckse@Nightstar-23b76b03.dsl.sentex.ca] has quit [Client closed the connection]
06:36 eckse [eckse@Nightstar-23b76b03.dsl.sentex.ca] has joined #code
06:36 mode/#code [+o eckse] by ChanServ
06:40 Kindamoody|gaming is now known as Kindamoody
06:58 Kindamoody is now known as Kindamoody|breakfast
07:00 Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Client closed the connection]
07:08 eckse_ [eckse@Nightstar-c957df70.dsl.sentex.ca] has joined #code
07:08 eckse [eckse@Nightstar-23b76b03.dsl.sentex.ca] has quit [Connection reset by peer]
07:09 eckse_ [eckse@Nightstar-c957df70.dsl.sentex.ca] has quit [Client closed the connection]
07:23 Vash [Vash@Nightstar-cdeba41f.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!]
07:28 Kindamoody|breakfast is now known as Kindamoody
07:35 Kindamoody is now known as Kindamoody|out
08:21 Kindamoody|out is now known as Kindamoody
09:07 Kindamoody [Kindamoody@Nightstar-5507a6b5.tbcn.telia.com] has quit [Ping timeout: 121 seconds]
09:11 Kindamoody|out [Kindamoody@Nightstar-5507a6b5.tbcn.telia.com] has joined #code
09:11 mode/#code [+o Kindamoody|out] by ChanServ
09:11 Kindamoody|out is now known as Kindamoody
09:13 You're now known as TheWatcher
09:33 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
09:46 himi [fow035@Nightstar-5d05bada.internode.on.net] has joined #code
09:46 mode/#code [+o himi] by ChanServ
10:32 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
11:06 maoranma [maoranma@Nightstar-c611b48e.pools.spcsdns.net] has quit [[NS] Quit: ]
11:34 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: Into the hole again, we hurried along our way, into a once-glorious garden now seeped in dark decay.]
11:39 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
11:40
< Stalker>
Ano of you have the necessary tools to open minidump files and are currently active?
11:44
<@jerith>
What are minidump files?
11:45 * jerith fiddles with his depixeling code, trying to do magic with integration.
11:47
<&McMartin>
Stalker: WinDbg should be a free download from MSDN
11:47
<&McMartin>
Actually *using* it I cannot teach you at 3:47 AM
11:48
<&McMartin>
But I can point you at http://windbg.info
11:48
<&McMartin>
http://msdn.microsoft.com/en-us/windows/hardware/gg463009
11:48
<&McMartin>
These open dump files
11:48
<&McMartin>
And, well
11:48
<&McMartin>
Good luck
11:49
< Stalker>
McMartin: I've come that far too, but I am not actually able to download it.
11:49
<&McMartin>
You can't get or run winsdk_web.exe?
11:49
<&McMartin>
It's a piece of what that installs.
11:50
< Stalker>
MSDN in general seems unwilling to provide a download button.
11:50
<&McMartin>
https://www.microsoft.com/download/en/details.aspx?displaylang=en&id=8279 ?
11:52
< Stalker>
That was easy, thank you.
11:52
<&McMartin>
Dunno why it didn't work for you at first, but there you go
11:52
<&McMartin>
It'll show up under "Debugging Tools for Windows" - make sure the x86/x64 part matches the minidump
11:53
<&McMartin>
windbg.info is full of tutorials but it's terrifyingly arcane overall
11:53
<&McMartin>
!analyze -v is a good start for BSOD minidumps though.
11:57
< Stalker>
Well, can't actually use this, since I have to boot in safe mode to avoid a BSOD.
11:57
< Stalker>
And safe mode does not have the windows installer, on which this relies.
11:57
< Stalker>
Thanks though, at least that exhausts that possebility.
11:58
<&McMartin>
If you're still having trouble tomorrow night, I might be able to try to look at it myself if you can put it somewhere
11:59
<&McMartin>
But I'm not likely to get you much more than the bluescreen itself will give you (that is, which module, and a stack trace with nothing you care about in it)
11:59 Attilla [Obsolete@Nightstar-839d06bf.as43234.net] has joined #code
11:59
<&McMartin>
Unless you've been developing your own device drivers or something.
11:59
<&McMartin>
You might try logging the boot to see what the last successful load is, maybe?
12:00
< Stalker>
Actually the cause of the bluescreen is a STOP 0xbe event, which is a driver trying to write in ROM memory.
12:00
< Stalker>
So I wanted to find out which driver it was so that when I wipe my box I can perhaps avoid the issue.
12:00
<&McMartin>
There should be a .sys or .dll listed too
12:00
<&McMartin>
As the one causing the STOP
12:01
< Stalker>
Oh?
12:01
< Stalker>
Didn't see one.
12:01
< Stalker>
But I'll try to provoke another.
12:01 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: Into the hole again, we hurried along our way, into a once-glorious garden now seeped in dark decay.]
12:01
<&McMartin>
It's right under the stop code
12:04 Rhamphoryncus [rhamph@Nightstar-5697f7e2.abhsia.telus.net] has joined #code
12:06 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
12:06
< Stalker>
Nope.
12:06 AnnoDomini [annodomini@A08927.B4421D.B81A91.464BAB] has joined #code
12:06 mode/#code [+o AnnoDomini] by ChanServ
12:07
<&McMartin>
Hm, odd
12:07
<&McMartin>
It should look like this:
12:07
<&McMartin>
http://www.maximumpc.com/files/u69/BSOD_Main.png
12:07
<&McMartin>
With a .sys under the STOP code
12:08
< Stalker>
It doesn't.
12:08
< Stalker>
No .sys line.
12:08
<&McMartin>
To get the faulting module or stack trace, !analyze -v from the dump should do it.
12:08
<&McMartin>
How big is the dump?
12:08
<&McMartin>
Also: x64 or x86?
12:09
< Stalker>
271 kb
12:09
< Stalker>
x64
12:09
<&McMartin>
Feel free to email it to mcmartin@gmail.com, I've got the x64 WinDbg installed right here
12:09
< Stalker>
Thank you.
12:10
<&McMartin>
No guarantees, though, if the BSOD didn't say anything itself
12:11
< Stalker>
Sent.
12:11
< Stalker>
Nah, that's fine.
12:12
< Stalker>
I'm going to wipe the box anyway, I just want to avoid future complications.
12:17
<&McMartin>
It's ntoskrnl.exe crashing, but I can't get the symbols to work right for me so I can't tell which driver it's dealing with, if any
12:17
< Stalker>
ntoskrnl sounds like something serious enough to actually warrant a reinstall.
12:17
< Stalker>
This happened before too, but then went away on its own.
12:17
< Stalker>
Now it's back, with a vengeance.
12:18
<&McMartin>
If you're getting through the credprov, and msconfig says nothing's there, the only other places to check would be the Startup folder, and the registry keys that hide additional autostart info
12:18
< Stalker>
At first I thought it was due to the frost freezing my RAM. (Frozen RAM attack - like)
12:18
< Stalker>
I've also disabled the registry keys.
12:18
<&McMartin>
You might want to take a look at HKCU\Software\Microsoft\Windows\CurrentVersion\Run
12:18
< Stalker>
I did.
12:18
<&McMartin>
And RunOnce, underneath it
12:19
<&McMartin>
And then the HKLM equivalents
12:19
<&McMartin>
k
12:19
<&McMartin>
memtest86 isn't actually too shabby an idea
12:19
< Stalker>
I had manually added a bat file to open some explorer windows which had been working fine. (This was a substitute to Explorer's native save-open-windows-thingie)
12:19
< Stalker>
Did that, no results.
12:19
< Stalker>
CHKdsk is also clean.
12:19
<&McMartin>
Yeah, you're getting through the login screen, it sounds like
12:20
< Stalker>
I am.
12:20
<&McMartin>
Which is crazy for BSODs, because once you're through the credential provider, damn near everything lives in userland
12:20
< Stalker>
I thought it might be explorer.exe being initialized too quickly (due previously mentioned bat) but the problem persisted through disabling all startup programs.
12:20
<&McMartin>
Also, this looks like it's Vista/7?
12:20
< Stalker>
7.
12:20
<&McMartin>
Explorer can survive termination in Vista/7.
12:21
< Stalker>
What do you mean?
12:21
<&McMartin>
Or rather
12:21
<&McMartin>
A login session can survive having explorer terminated
12:21
< Stalker>
Been able to do that since 95.
12:21
<&McMartin>
Somebody else - I think it's LogonUI.exe - will restart it for you.
12:21
<&McMartin>
Right, but the thing is, explorer also lives in userspace
12:21
< Stalker>
.. well, I've been terminating explorer.exe whenever there's explorer problems since 95 at least.
12:22
<&McMartin>
It shouldn't be able to bluescreen you under any circumstances
12:22 * Stalker shrugs.
12:22
<&McMartin>
ISTR in 95 and 98 you had to restart it by hand afterwards
12:22
<&McMartin>
through the task manager or whatever
12:22
< Stalker>
At this point I was just trying to secure everything.
12:22
<&McMartin>
Don't remember for XP
12:22
<&McMartin>
Yeah
12:22
< Stalker>
Still do for XP.
12:22
<&McMartin>
I'd say "back up important data, then do your nuke and pave, see what's up"
12:22
< Stalker>
Did the backuping already.
12:22
<&McMartin>
By the time you hit the login GUI *at all* all device drivers should be completely loaded. =/
12:23
< Stalker>
Yup.
12:23
< Stalker>
It idles just fine at the login GUI too.
12:23
<&McMartin>
Though I guess it wasn't a driver crashing, huh, it was ntoskrnl.exe~
12:23
< Stalker>
For at least 10 minutes.
12:23
< Stalker>
Possibly.
12:24
<&McMartin>
Yeah, I guess I should say "that's the module the dump identifies"
12:24
< Stalker>
Oh well, see you on the other side, I guess.
12:24
<&McMartin>
Good luck
12:24
< Stalker>
Thanks.
12:24 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: Into the hole again, we hurried along our way, into a once-glorious garden now seeped in dark decay.]
12:41
< Rhamphoryncus>
#define DECLARE_POSTFIX_INCREMENT(type) \
12:41
< Rhamphoryncus>
inline type operator ++(type& e, int) \
12:41 * Rhamphoryncus develops a twitch
13:12 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
13:23 Red_Queen [NSwebIRC@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
13:23
< Red_Queen>
Well, that seems to have worked.
13:23
< Red_Queen>
Now I'm updating my windoooze.
13:36
< Red_Queen>
Thank you McM.
13:47 gnolam [lenin@Nightstar-202a5047.priv.bahnhof.se] has joined #code
14:43 Red_Queen [NSwebIRC@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: Page closed]
14:56 Thalass [Thalass@Nightstar-70faba70.bigpond.net.au] has joined #code
14:59
< Thalass>
Stupid question: I'm reinstalling ubuntu and want / and /home to be separate partitions. How large does / have to be?
14:59
<@Tamber>
Not terribly.
15:00 * Tamber has / at 40G, and is only using 21G of that.
15:00
<@Tamber>
(And that's with the source of everything, including LOffice.)
15:02
< Thalass>
Thanks. I'm checking how large the backup of my /home is via the wide's laptop. Should have checked before starting. :P
15:02
<@Tamber>
:p
15:07 Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code
15:07 mode/#code [+o Syloqs_AFH] by ChanServ
15:08 Syloqs_AFH is now known as Syloqs-AFH
15:19
< Thalass>
Righty. 50gb for /, 194gb for /home, and 5gb for swap. Then spare at the end because i'm going to try and get expressgate working again.
15:19
< Thalass>
Not that i'll use it much haha
15:27
<@ToxicFrog>
Thalass: I've got 20GB for / here and it's less than half used.
15:27
<@ToxicFrog>
50GB is almost certainly a lot of overkill.
15:28
<@ToxicFrog>
What's an expressgate?
15:28
< Thalass>
dangit i just clicked install haha
15:29
< Thalass>
Asus netbooks have a second power button that quickly boots a minimalist OS that runs firefox, some kind of IRC, and a couple of other netbook related things.
15:31 Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code
15:31
< Thalass>
Oh well, this install only has to last until i get back from Canada. By then the new Ubuntu should be out and i can decide which flavour of linux to go with.
15:33 Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
15:34 Syloqs-AFH [Syloq@NetworkAdministrator.Nightstar.Net] has joined #code
15:35
<@ToxicFrog>
Well, you can always use gparted to resize the partitions if you need to
15:35 * Thalass nods
15:36 Syloqs_AFH [Syloq@NetworkAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
15:36 Syloqs-AFH is now known as Syloqs_AFH
15:36 mode/#code [+o Syloqs_AFH] by ChanServ
15:42
< Rhamphoryncus>
Woot, I think my timetabling is actually reliable this time
15:43
<@Tamber>
Now you've doomed it.
15:43
< Rhamphoryncus>
Even when I turned on breakdowns and a bunch of them in a row through it off.. after a minute it recovered. Briefly, before more breakdowns through it off.
15:46 eckse [eckse@Nightstar-086443b9.dsl.sentex.ca] has joined #code
15:46 mode/#code [+o eckse] by ChanServ
15:52
< Thalass>
Is this OTTD?
15:53
< Thalass>
(dangit. Fresh install and the brightness buttons are wonky again. I'd forgotten about that. )
16:01
< Rhamphoryncus>
yes
16:01
< Rhamphoryncus>
I've been hacking on the timetabling for the last 2-3 weeks now
16:09
< Thalass>
neat. I haven't really used timetables much, yet.
16:14
< Rhamphoryncus>
They're garbage
16:15
< Rhamphoryncus>
Each vehicle is totally independent from each other, so if one is late it has to make up the lost time all by itself
16:17
< Rhamphoryncus>
Start times have to be manually set for each vehicle, using a calendar day. 5 vehicles on a 44 day cycle? 9, 9, 9, 9, 8, which means.. febuary 1st, febuary 10th, febuary 19th, febuary 28th, march 7th
16:17
< Rhamphoryncus>
And the kicker, various things (such as sending your vehicle to the depot) can break synchronization
16:20
< Thalass>
ah. Glad i haven't, then
16:21
< Thalass>
I'm still hanging out for cargodest and subsidiaries returning.
16:21
< Rhamphoryncus>
yeah
16:28 Thalass [Thalass@Nightstar-70faba70.bigpond.net.au] has quit [[NS] Quit: phone off]
16:31 thalass [thalass@Nightstar-70faba70.bigpond.net.au] has joined #code
16:31
< thalass>
ah. Much better.
16:32
< thalass>
Hrm. Brightness issue on my netbook. Updating the bios is recommended. I didn't do that last time.
16:39
< Rhamphoryncus>
Bah. Now I have a new circumstance I dislike. Not late at all :P
16:41
< Rhamphoryncus>
More specifically, take a route with 4 trains and 4 stops, equally spaced. As long as there isn't too much slack they'll balance out and become synchronized.
16:42 Vash [Vash@Nightstar-cdeba41f.wlfrct.sbcglobal.net] has joined #code
16:42 mode/#code [+o Vash] by ChanServ
16:42
< Rhamphoryncus>
But take that same route and boost it to 8 trains... now their wait times overlap so you have much more slack, meaning you can have 3 trains at one stop and 1 at another and it's still not late
16:43
< Rhamphoryncus>
So.. the first there is to have a *third* criteria for idle vehicles
16:46
<~Vornicus>
Is this a circle route or a back and forth route?
16:46
< Rhamphoryncus>
Or say fuck it and let the user prod them if they want it to be pretty
16:46
< Rhamphoryncus>
circle
16:47
< Rhamphoryncus>
I think the underlying problem is when creating a route you take the average transit+loading time and add some slack, but when increasing the number of vehicles the slack shouldn't scale up as much
16:48
<~Vornicus>
With 8 trains on a 4-station route like that you should have 4 stopped and 4 moving at any given time.
16:48
< Rhamphoryncus>
That's my expectation
16:50
<~Vornicus>
Which I think in all situations you can get by doing "go if a clone arrives"
16:51
< Rhamphoryncus>
That'd be a special case. I'm talking general here.
16:52 thalass [thalass@Nightstar-70faba70.bigpond.net.au] has quit [[NS] Quit: for great justice and sleep]
16:55
< Rhamphoryncus>
aha, the disparate loading times are what's throwing it off
16:57
< Rhamphoryncus>
First vehicle that arrives gets first dibs on whatever there is to load. It's possible if they're slow to load and unlucky to stay loading keep past their departure time because more keeps showing up
16:58
< Rhamphoryncus>
But I consider them already departed
16:59
<~Vornicus>
So if they're supposed to depart, but more stuff keeps showing up to get loaded before it finishes loading what it's got, then you're screwed.
16:59
< Rhamphoryncus>
yup
17:00
<~Vornicus>
What does your considering them already departed do?
17:00
< Rhamphoryncus>
But then the next vehicle doesn't do that, so they leave in a burst
17:00
< Rhamphoryncus>
It means they're not late
17:00
< Rhamphoryncus>
At least not at that station
17:00
< Rhamphoryncus>
And because there are extra trains they're not likely to be late at the next one
17:01
< Rhamphoryncus>
I do occasionally see a late arrival, but it's too rare to trigger a skip
17:02
< Rhamphoryncus>
It seems as if delivering passengers causes the game to produce more.. which might very well be the case
17:03
< Rhamphoryncus>
Oh hey, I just saw a train overtake another in a station
17:03
< Rhamphoryncus>
It took so long to unload that the following train unloaded first
17:11
< Rhamphoryncus>
hmm, actually, I think I can fix this. I have a flag indicating a train is waiting for a departure. If I have a departure and that flag is set it means another train is waiting.
17:35
< Rhamphoryncus>
Vehicle 103 running away!
17:35
< Rhamphoryncus>
:D
17:36
<@Tamber>
:D
17:37
<~Vornicus>
Is that a quote?
17:38
< Rhamphoryncus>
nope. Just my debugging print
17:40
< Rhamphoryncus>
But it may harken back to this somehow: http://archive.kontek.net/rott.classicgaming.gamespy.com/fun/imfree.jpg
17:58 Ling is now known as Anna
18:13 * AnnoDomini lols at job advert. They want developer nerds, but put in bold letters that good personal hygiene is required.
18:17 celticminstrel [celticminst@Nightstar-5d22ab1d.cable.rogers.com] has joined #code
18:34 Derakon[AFK] is now known as Derakon
18:38
< Rhamphoryncus>
Problem fixed without the running away thing. It was a regression due to my patch.
18:39
< Rhamphoryncus>
If a vehicle loaded only a partial amount in the last tick, meaning not at the maximum rate they can, the timetable is supposed to take precedence. Unfortunately I didn't update the bit of code looking at the timetable. :)
18:41 maoranma [maoranma@Nightstar-c611b48e.pools.spcsdns.net] has joined #code
18:41
<@AnnoDomini>
maoranma: First bugfix released today.
18:42
< maoranma>
Saw
18:42
<@AnnoDomini>
Wrench.
18:42
< maoranma>
Bet adv mode still hard as fuck
18:43 maoranma [maoranma@Nightstar-c611b48e.pools.spcsdns.net] has quit [[NS] Quit: ]
18:44 maoranma [maoranma@Nightstar-c611b48e.pools.spcsdns.net] has joined #code
18:45
< maoranma>
Because I'm tired of getting shot in the brain by crosseyed kobold #3
18:58 Kindamoody is now known as Kindamoody[zZz]
18:59 EvilDarkLord is now known as Aeron
19:01 AnnoDomini is now known as Jasever
19:09
< Rhamphoryncus>
https://bitbucket.org/rhamph/openttd-rhamph/changesets/tip/branch%28%22route%22% 29
19:24 eckse [eckse@Nightstar-086443b9.dsl.sentex.ca] has quit [Ping timeout: 121 seconds]
19:29 eckse [eckse@Nightstar-086443b9.dsl.sentex.ca] has joined #code
19:29 mode/#code [+o eckse] by ChanServ
19:39 Red_Queen [NSwebIRC@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
19:40
< Red_Queen>
So far, so good. Almost done patching windows (I hope). Could anyone explain to me why it refuses to acknowledge the existance of my external HDD?
19:48
<@ToxicFrog>
As in, you plug the disk in and it isn't assigned a drive letter?
19:48
<@ToxicFrog>
Or as in you plug it in and it doesn't even appear in the Disk Manager?
19:51
< Red_Queen>
As in I plug in the disk, and both the device manager recognizes something is plugged in, and the hdd begins blinking in activity, but that's all that happens.
19:52
< Red_Queen>
The last five windows update updates also refuses to download.
19:52
<@ToxicFrog>
Open disk manager, see if the disk shows up there and what its partition layout is
19:53
<@ToxicFrog>
It might be having a crazy and assigning drive letters that are already in use (this seems to be very common when networked drives are being used), or the disk might not have any filesystems on it that windows knows how to mount.
19:53
< Red_Queen>
This disk worked fine before I wiped and reinstalled my computer.
19:53
< Red_Queen>
It has my backup.
19:53
<@ToxicFrog>
(disk manager is under Control Panel -> Administration somewhere, IIRC)
19:53
< Red_Queen>
Yeah, computer management.
19:54
< Red_Queen>
Does not show up.
19:54
< Red_Queen>
I assume it's because Windows does not actually recognize it as an external HDD:
19:54
< Red_Queen>
Even though the device manager writes "USB Mass Storage device."
19:55
<@ToxicFrog>
That's weird. I've got nothing.
19:55
< Red_Queen>
Windows has determined the driver software is up to date >_<
19:57
<@ToxicFrog>
Good luck.
19:58
< Red_Queen>
Aha, here we go.
19:58
< Red_Queen>
Apparently windows had decided to auto-disable it in the device manager.
20:07
< Red_Queen>
GERONIMO!
20:08
< Red_Queen>
Well, that failed spectacularly.
20:10
< Red_Queen>
Round 2 - FIGHT!
20:10
< Red_Queen>
Hmm, my settings are borked.
20:15 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
20:19
< Stalker>
Something is decidedly wrong.
20:20 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: ]
20:22
< Eri>
Man, I have an intrinsic fuzzy-filter on this crt monitor.
20:22
< Eri>
Really should replace it, someday.
20:22
< Eri>
I was looking up some mysql stuff, and I saw vorns-mysql-plugin, and was wondering if it was our vorn
20:22
< Eri>
Then I put it on the other monitor, and realized it was voms
20:26 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has joined #code
20:26
< Stalker>
Nice, thanks guys.
20:27 Stalker [Z@Nightstar-3602cf5a.cust.comxnet.dk] has quit [[NS] Quit: ]
20:31 Red_Queen [NSwebIRC@Nightstar-3602cf5a.cust.comxnet.dk] has quit [Ping timeout: 121 seconds]
21:28 RichyB [MyCatVerbs@Nightstar-4764abfc.bb.sky.com] has joined #code
22:10 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has quit [Connection reset by peer]
22:26 ToxicFrog [ToxicFrog@ServerAdministrator.Nightstar.Net] has joined #code
22:26 mode/#code [+o ToxicFrog] by ChanServ
22:30 RichyB [MyCatVerbs@Nightstar-4764abfc.bb.sky.com] has quit [Ping timeout: 121 seconds]
22:33 RichyB [MyCatVerbs@Nightstar-4764abfc.bb.sky.com] has joined #code
23:11 Aeron is now known as EvilDarkLord
23:27 Anna is now known as Ling
23:29 Jasever is now known as AnnoDomini
23:51 AnnoDomini [annodomini@A08927.B4421D.B81A91.464BAB] has quit [[NS] Quit: leaving]
--- Log closed Sun Feb 19 00:00:36 2012
code logs -> 2012 -> Sat, 18 Feb 2012< code.20120217.log - code.20120219.log >

[ Latest log file ]