code logs -> 2012 -> Thu, 30 Aug 2012< code.20120829.log - code.20120831.log >
--- Log opened Thu Aug 30 00:00:15 2012
00:23 * Vornicus pokes at No Stomach For Lunch, determines he has no idea yet how to do it.
00:25
<~Vornicus>
I can certainly drag stuff around to make the ethane happen, but the hydrogens steadfastly refuse to be put back in their places.
00:26
< ToxicFrog>
Which one is that again?
00:26
< ToxicFrog>
I really need to get back into, and at some point finish, Spacechem
00:26
<&McMartin>
XOTHOTHOR
00:26
<&McMartin>
(XOTHOTHOR!)
00:26
<&McMartin>
HE'S IMMUNE TO BULLETS
00:26
<&McMartin>
AND LOVE
00:27
<~Vornicus>
7-2: Ethane to Ethene and Hydrogen, in a micro bonding reactor, flip flops are available.
00:27
<~Vornicus>
I think I might see a trick here actually.
00:27
<~Vornicus>
...no, that doesn't work, the bond can't happen until the second hydrogen is off.
00:28
< ToxicFrog>
McMartin: I don't remember which one XOTHOTHOR is either~
00:28
<&McMartin>
Pretty sure you didn't reach him.
00:28
<&McMartin>
XOTHOTHOR is the name of the boss in Chapter 8, IIRC.
00:29
< ToxicFrog>
My latest boss is BAH SHOGTH, THE CONTROLLER
00:29
< ToxicFrog>
Which is chapter 7.
00:29
< ToxicFrog>
Also, wow, I haven't optimized No Stomach for Lunch at all
00:29
< ToxicFrog>
415 clocks, 38 opcodes.
00:30 * Vornicus is vaguely trying to speed up his version of The Blue Danube, just one step shorter somewhere would make it work.
00:30
<&jerith>
Idly, http://www.kongregate.com/games/jahooma/jahoomas-logicbox
00:31
<&jerith>
Number 12 gave me a little trouble until I hit on the right approach.
00:31
<&jerith>
The rest were fairly straightforward.
00:32 * ToxicFrog ponders The Blue Danube. Why did I do it that way?
00:36
<~Vornicus>
I don't know, but yours is faster than mine.
00:37
< ToxicFrog>
I wish spacechem had semaphores as well as barriers.
00:41
< ToxicFrog>
...I think I see a way to do this without FFs.
00:41
< ToxicFrog>
I'm not sure it'd actually be faster, though.
00:43
<~Vornicus>
The Blue Danube is The Danger Zone, except with Flipflops
00:44
<~Vornicus>
My flipflop version is significantly faster.
00:44
< ToxicFrog>
How the hell did childhood flu symptoms get this down to 165
00:45
<~Vornicus>
childhood flu symptoms?
00:45
< ToxicFrog>
Someone on my friends list
00:45
< ToxicFrog>
That's an average of 8.25 clocks/molecule
00:47
<~Vornicus>
...at 27-28 inputs, that's under 6 clocks per input molecule. That's recockulous.
00:47
< ToxicFrog>
Yes.
00:47
<~Vornicus>
...hang on,
00:47
<~Vornicus>
I bet you this is a parallelized version.
00:47
< ToxicFrog>
Well, yes, of course
00:47
< ToxicFrog>
12 clocks per molecule per thread
00:48 You're now known as TheWatcher[T-2]
00:50
< ToxicFrog>
I really wish you could save entire solutions >.<
00:50 You're now known as TheWatcher[zZzZ]
00:52
<~Vornicus>
Me too.
01:00
<~Vornicus>
221...
01:00
< ToxicFrog>
I also really wish you could invert sensors.
01:00
<~Vornicus>
Like make them go "not this" instead of "this"?
01:00
< ToxicFrog>
Yes.
01:01
<&McMartin>
Can't you just point the arrow forward and include a directional arrow in that spot?
01:01
< Tarinaky>
How do you do mouse picking in 3D?
01:01
<&McMartin>
It's a fair amount of work, actually.
01:01
<&McMartin>
OpenGL 1.0 had some routines just for this and I think it forwarded it to the rasterizer
01:01
< gnolam>
Tarinaky: various methods.
01:02
<~Vornicus>
McM: sure, but sometimes you want to do crazy stuff.
01:02
< gnolam>
There's actual picking: rendering the scene, each object a different color, and checking the pixel in the mouse's hot zone.
01:02
< ToxicFrog>
McMartin: upon considering, this may have the same semantics and in either case it doesn't actually do what I want ;.;
01:02
< gnolam>
Then there's regular raycasting.
01:02
<~Vornicus>
There are some subtleties in the semantics here though
01:03
<~Vornicus>
Consider: you can reach the sensor from a variety of directions, but activating it forces you in one direction.
01:03
<~Vornicus>
Not activating it, you continue in whatever direction you were going.
01:04
< Tarinaky>
Is raycasting the more 'correct' way?
01:05
< gnolam>
Depends on the application really, but in general yes - for the simple reason that it keeps logic and rendering separate.
01:05 himi [fow035@Nightstar-5d05bada.internode.on.net] has quit [Ping timeout: 121 seconds]
01:06
<&McMartin>
Vornicus: I'm not convinced that makes topological sense without three dimensions
01:06
< gnolam>
But as always, that's a guideline and not a strict rule.
01:06
< Tarinaky>
Where would I find out about raycasting?
01:06
<~Vornicus>
McM: to be honest I'm a little befuddled where I'd use that myself.
01:08
< Tarinaky>
I've not seen it mentioned in the SuperBible.
01:08
< Tarinaky>
Although I admit I've read none of them cover to cover (yet)
01:15 * Vornicus pokes at jahooma's logic box, gets stuck on #6.
01:16
<~Vornicus>
I can't figure out how to tell if there's only one symbol left.
01:17
<~Vornicus>
oh there it goes.
01:18 * Vornicus turns from a bicycle into a ....cactus.
02:14 Attilla [Obsolete@Nightstar-8e1acbff.as43234.net] has quit [Ping timeout: 121 seconds]
02:16
< ToxicFrog>
I still kind of want to write a Spacechem compiler sometime.
02:19
<~Vornicus>
what would it do?
02:21
< ToxicFrog>
Take input in some as-yet-to-be-determined source form and output Spacechem reactor complex designs.
02:21
<~Vornicus>
...yeesh.
02:26 * Vornicus finally manages to shave the 2 cycles he needs off of Sleepless on Sernimir IV to tie with TF's speed number there.
02:28
<~Vornicus>
Next one I need to optimize to get into first place again: Settling Into The Routine, I need to somehow find a 10-fewer-symbols solution.
02:29
< ToxicFrog>
I don't even bother to optimize for code size, as a rule; I don't find it interesting.
02:29
<~Vornicus>
I find all three interesting.
02:32 * Vornicus reduces symbol count by 11 instead.
02:33 * Vornicus thereby defeats iospace's solution. :P
02:33
< iospace>
:P
02:39
<~Vornicus>
Next: delete 2+ symbols from Nothing Works.
02:57 himi [fow035@D741F1.243F35.CADC30.81D435] has joined #code
02:57 mode/#code [+o himi] by ChanServ
03:15
<~Vornicus>
gnerk. my previous method was so fast I was at streaming and had no syncs across the entire board.
03:17
< ToxicFrog>
It's always nice when you can do that.
03:17
<&McMartin>
I optimized for programmer time
03:18
<&McMartin>
This means I can never revisit them
03:18
< ToxicFrog>
Spacechem really needs integrated version control, dammit
03:18
< ToxicFrog>
I'm kind of tempted to write some.
03:18
< ToxicFrog>
The save files are just sqlite3 databases, after all.
03:24
<@Alek>
har har har, mcm
03:34
<~Vornicus>
I keep kind of wishing for a more detailed scores screen.
04:17
<@Alek>
and a way to save/compare solutions. >_>
04:23 Derakon[AFK] is now known as Derakon
04:23 mode/#code [+ao Derakon Derakon] by ChanServ
04:26 * ToxicFrog cuts 1000 cycles off Right All Along, is still 700 cycles too slow for the #1 spot
04:33
<~Vornicus>
Achievement Unlocked: Cost Control Specialist. Falling done in two reactors.
04:44 * Vornicus still befuddles at TF's speed in The Plot Thickens
04:47 Kindamoody[zZz] is now known as Kindamoody
04:57
< ToxicFrog>
:D
04:57
< ToxicFrog>
Sadly I'm not as impressive in Right All Along, which is the FF equivalent.
04:57 Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has quit [Ping timeout: 121 seconds]
04:58 Derakon [Derakon@Nightstar-a3b183ae.ca.comcast.net] has joined #code
04:58 mode/#code [+ao Derakon Derakon] by ChanServ
05:21 iospace is now known as iospacedout
05:22 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has quit [Ping timeout: 121 seconds]
05:35 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has joined #code
05:43
<~Vornicus>
I'm still trying to figure out how you did it, that's under ten per monomer.
05:54 celticminstrel [celticminst@Nightstar-05d23b97.cable.rogers.com] has quit [[NS] Quit: And lo! The computer falls into a deep sleep, to awake again some other day!]
05:57 Kindamoody is now known as Kindamoody|out
06:29
<&McMartin>
Ahahahahaha.
06:29
<&McMartin>
That is *excellent*
06:29
<&McMartin>
And maybe 12 people will care
06:29
<&McMartin>
But I am one of them!
06:29
<&McMartin>
So, as you know Bob (tm), BIT.TRIP runner has some shoutouts to the Atari 2600 generally and Activision's games for it in particular.
06:30
<&McMartin>
It's only really noticable if you do it in windowed mode, but they also simulated the side effect of Activision's systematic abuse of the 2600's HMOVE register.
06:30
<~Vornicus>
namely?
06:30
<~Vornicus>
(also that's some esoterica there)
06:31
<&McMartin>
Bar of 8[*] black pixels down the left side of the screen.
06:31 Derakon is now known as Derakon[AFK]
06:31
<&McMartin>
[* on a 320-pixel display]
06:31
<~Vornicus>
aha
06:33
<&McMartin>
If you weren't systematic about it, the left edge would start at different points, producing a ragged left edge. Air/Sea Battle was the most flagrant about this.
06:33
<&McMartin>
I learned about Activision's penchant for this from Nick Montfort's book "Racing The Beam", and connected it to the ragged-edge stuff personally by reading the disassembly of A/SB.
06:36
<~Vornicus>
what I remember about A/SB is how between each row of critters there was a thin line at the left side.
06:37 cpux|2 [cpux@Nightstar-c5874a39.dyn.optonline.net] has joined #code
06:39 cpux [cpux@Nightstar-c5874a39.dyn.optonline.net] has quit [Ping timeout: 121 seconds]
06:40
<&McMartin>
Yup.
06:40
<&McMartin>
That would happen if you did the trickery to get more than two sprites on screen at once.
06:40
<&McMartin>
So Activision would hide that by doing the trickery every single line, even if there was nothing to do.
06:40
<&McMartin>
Putting that "thin line" on every single scanline.
06:40
<&McMartin>
The bonus stage for RUNNER puts that black bar on the left side of the window, but only during the 2600-y bonus stages.
06:46
<~Vornicus>
Man, I kinda miss some of those atari games.
06:46
<~Vornicus>
Though to be honest most of them were kind of shit
06:46
<~Vornicus>
more than kind of.
06:51
<~Vornicus>
oh, there was the kangaroo one with the hilariously picky jumping physics, and the one where it was a ghost house one and something else, a 2-in-1 cartridge...
06:52
<~Vornicus>
and hall of the mountain king
06:52
<~Vornicus>
and...
06:52
<~Vornicus>
and then I think about other things that came out a little later and they honestly weren't much better. Gogo the Ghost.
06:55
<~Vornicus>
And now I have the Pitfall II song stuck in my head.
07:06
< Rhamphoryncus>
Anybody know of a convention on C++ factory functions? Specifically wrappers for the normal constructors, but with less inane limitations
07:08
< Rhamphoryncus>
More specifically, I want my Matrix class to be constructible using a begin and end iterator
07:08
<~Vornicus>
When you say "convention", what flavor of convention do you wish?
07:09
< Rhamphoryncus>
It's perfectly legal to declare it as such, but you can't explicitly give the type of the iterator, nor is it willing to infer it
07:09
<&McMartin>
Two conventions
07:09
<&McMartin>
(1) factory functions go in the same namespace but outside of the class
07:09
<&McMartin>
(2) factory function is a static within the class.
07:10
< Rhamphoryncus>
Foo x = Foo(first, last); // not allowed, so I have to do this: Foo x = Foo::make(first, last);
07:11
< Rhamphoryncus>
ergh. Question is a *naming* convention
07:12
<&McMartin>
Foo::make, Foo::create are both fine
07:12
<&McMartin>
out of class, make_foo, new_foo, and create_foo all fine.
07:12 * Rhamphoryncus nods
07:14
<~Vornicus>
I've occasionally seen stuff like Complex is the class and then outside the class I've seen PolarComplex
07:14
< Rhamphoryncus>
I've also ran into this with immutable instances. You have to pass in the values essentially intact, or otherwise have a helper function for each to fix them up
07:44
< Rhamphoryncus>
template<class T> static Matrix<N> make(decltype(T::begin()), decltype(T::end()));
07:45
< Rhamphoryncus>
It "accepts" that but it's not the right way of getting the type for begin/end. Any idea what is the right type?
08:14 Vornotron [vorn@ServerAdministrator.Nightstar.Net] has joined #code
08:17 Vornicus [vorn@ServerAdministrator.Nightstar.Net] has quit [Ping timeout: 121 seconds]
08:19 Vash [Vash@Nightstar-e8057de2.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!]
09:11
< Rhamphoryncus>
Fixed, apparently:
09:11
< Rhamphoryncus>
template<class T> static Matrix<N> make(decltype(std::declval<T>().begin()), decltype(std::declval<T>().end()));
09:37 himi [fow035@D741F1.243F35.CADC30.81D435] has quit [Ping timeout: 121 seconds]
09:48 You're now known as TheWatcher
11:43 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has quit [Client exited]
11:47 * AnnoDomini has purchased a can of air!
11:47
< AnnoDomini>
Time to clean my laptop a bit. I wonder if doing this while in operation will crash it.
11:47
< AnnoDomini>
Let's find out!
11:48
<@TheWatcher>
I generally recommend against it, if you're applying it anywhere likely to involve fans
11:49
<@TheWatcher>
The air blast may force the fans to rotate the wrong way, and arse up the sense signal to the controller.
11:50
< AnnoDomini>
I see.
12:01
<@TheWatcher>
(not serious, but I have crashed a couple of laptops doing it)
12:02 RichyB [richardb@Nightstar-428e5766.bb.sky.com] has joined #code
12:03 Attilla [Obsolete@Nightstar-8e1acbff.as43234.net] has joined #code
12:12
< AnnoDomini>
Hmm. If I take out the battery out of a Dell Inspiron 1525 when it is plugged into the grid and operating, will its power fail?
12:20
<@Tamber>
Shouldn't. But then, it *is* a Dell, so it could do anything from "work perfectly" to "explode and destroy all you hold dear"!
12:40 gnolaptop [lenin@Nightstar-47fbf150.eduroam.liu.se] has joined #code
12:44 RichyB [richardb@Nightstar-428e5766.bb.sky.com] has quit [[NS] Quit: Leaving]
12:44 * Alek misses Warlord. Atari 4-player breakout. :P
12:46
<@TheWatcher>
Argh, goddamned language clashes.
12:46 * TheWatcher removes the leading $ from variables in this javascript
12:49 Nemu [NeophoxProd@Nightstar-d60f94b0.asahi-net.or.jp] has joined #code
12:59 * gnolaptop points and laughs at TheWatcher.
12:59
< gnolaptop>
(But see also: semicolons.)
13:38 McMartin_ [mcmartin@Nightstar-b7749e07.pltn13.sbcglobal.net] has joined #code
13:41 McMartin [mcmartin@Nightstar-62073b2a.pltn13.sbcglobal.net] has quit [Ping timeout: 121 seconds]
13:57 Kindamoody|out is now known as Kindamoody
14:00 gnolaptop [lenin@Nightstar-47fbf150.eduroam.liu.se] has quit [[NS] Quit: Gone]
14:28 celticminstrel [celticminst@Nightstar-05d23b97.cable.rogers.com] has joined #code
14:40 iospacedout is now known as iospace
14:46 * TheWatcher flails at circular reference problems
15:06 gnolam [lenin@Nightstar-ccbf4b44.cust.bredbandsbolaget.se] has quit [Client closed the connection]
15:06 gnolam [lenin@Nightstar-ccbf4b44.cust.bredbandsbolaget.se] has joined #code
15:33 * TheWatcher facepalms, realises that he'd already built-in a circular reference cleanup system, he just needed to set up this module to use it
15:46
< iospace>
:P
16:20 You're now known as TheWatcher[afk]
16:59 * Vornotron fiddles with No Ordinary Headache.
16:59
< Vornotron>
...826, djord. Really.
17:00
< AnnoDomini>
Is it bad that aptitude reports: "Current status: 0 broken [-1]."?
17:01
< Vornotron>
pffff
17:02
< AnnoDomini>
I think it's been that way for a while.
17:17 celticminstrel is now known as celmin|away
17:20 Vornotron is now known as Vornicus
17:20 mode/#code [+qo Vornicus Vornicus] by ChanServ
18:06 Kindamoody is now known as Kindamoody|movie
18:19 Vash [Vash@Nightstar-e8057de2.wlfrct.sbcglobal.net] has joined #code
18:20 mode/#code [+o Vash] by ChanServ
18:46
< Namegduf>
AnnoDomini: I don't know what your complaint is, it's clearly so resilent that it's got -1 broken packages.
18:49
< AnnoDomini>
I have no complaint. Things are working fine. I just want to know what that -1 is doing there. :p
19:18 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has joined #code
20:11 * AnnoDomini hugs Gramps.
20:11
< AnnoDomini>
It is a nice program.
20:12
<@Tamber>
It just smells a littly funny, and constantly tries to feed you Werthers Originals?
20:14
< AnnoDomini>
Something like that. :P
20:14 * AnnoDomini likes Werthers Originals.
20:15
<@Tamber>
:)
20:16
< McMartin_>
Man, this code. -_-
20:18
<~Vornicus>
how -_- is it.
20:18
< McMartin_>
This remains to be determined.
20:19
< McMartin_>
I'm trying to track where user-supplied information is coming from when the core uses it
20:19
< McMartin_>
Whoever wrote this was Very Enterprisey Indeed.
20:19
< McMartin_>
I'm still drilling through wrappers.
20:23
< McMartin_>
OK.
20:24
< McMartin_>
I count four levels of object containment here.
20:52
<&jerith>
McMartin_: You are enthwipped.
21:05 * Vornicus thinks "ent-whipped"?
21:09
<&jerith>
That tail is called a "thwip".
21:11 celmin|away [celticminst@Nightstar-05d23b97.cable.rogers.com] has quit [Client closed the connection]
21:11 gnolam [lenin@Nightstar-ccbf4b44.cust.bredbandsbolaget.se] has quit [Client closed the connection]
21:11 celticminstrel [celticminst@Nightstar-05d23b97.cable.rogers.com] has joined #code
21:11 gnolam [lenin@Nightstar-ccbf4b44.cust.bredbandsbolaget.se] has joined #code
21:15 Tamber [tamber@furryhelix.co.uk] has quit [Connection closed]
21:15 Tamber [tamber@furryhelix.co.uk] has joined #code
21:15 mode/#code [+o Tamber] by ChanServ
21:18 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has quit [Ping timeout: 121 seconds]
21:24 McMartin_ is now known as McMartin
21:24 mode/#code [+ao McMartin McMartin] by ChanServ
21:24 Rhamphoryncus [rhamph@Nightstar-cc6253d6.abhsia.telus.net] has joined #code
21:25 You're now known as TheWatcher
21:26 Vash [Vash@Nightstar-e8057de2.wlfrct.sbcglobal.net] has quit [[NS] Quit: I lovecraft Vorn!]
21:37 Kindamoody|movie is now known as Kindamoody
21:57
< AnnoDomini>
Why does my can of compressed air get cold when I use it?
21:57
< AnnoDomini>
I think the answer is PHYSICS but why exactly?
21:58
< Namegduf>
When something pressurised ceases to be pressurised it gets colder.
21:58
< Namegduf>
I don't know more detail than that.
22:02
< ToxicFrog>
"combined gas law" may be a useful search key here.
22:03
< AnnoDomini>
Hmm. I know that the reverse is true: if something is pressed, it gets hotter. I guess the reverse is also true.
22:03
< AnnoDomini>
+grammatical sense
22:04
<~Vornicus>
For compressed air it goes something like this: you have a thing that reduces the amount of gas in the can, but the can's volume doesn't change, and the pressure also doesn't change enough to offset the loss of mass.
22:04
<~Vornicus>
...did I get my signs confused again?
22:05
< ToxicFrog>
Mass is irrelevant here, I think - just volume, pressure, and temperature.
22:05
<~Vornicus>
It's also "amount of gas"
22:06
< ToxicFrog>
pV/T is constant; V has remained the same, and p has decreased, so T must also decrease.
22:06
<~Vornicus>
PV = nRT, pressure * volume = moles of gas (mass equivalent) * ideal gas constant * temperature.
22:06
< ToxicFrog>
(more generally, yes, compressing something heats it up and decompressing it cools it down)
22:07
<~Vornicus>
The amount of gas in the can isn't contant.
22:07
<~Vornicus>
--wait, compressed air actually has a different reason,
22:08
<~Vornicus>
Inside the canned air is actually a thing that's /liquid/ at the pressures involved.
22:08
<~Vornicus>
liquid -> gas requires the use of energy, so the reaction is endothermic.
22:11
< ToxicFrog>
Oh seriously? I did not know that.
22:11
< ToxicFrog>
That it was liquid in the can, I mean.
22:12
<~Vornicus>
http://en.wikipedia.org/wiki/Gas_duster
22:12
<~Vornicus>
This is why you have to use it upright. It spits otherwise.
22:24
< celticminstrel>
I see I disconnected while I was away.
22:24 Kindamoody is now known as Kindamoody[zZz]
23:33
< Rhamphoryncus>
I thought it was something like temperature being proportionate to pressure and thus to volume
23:38
< Rhamphoryncus>
Although.. naively, temperature would influence pressure but not the other way around (an individual molecule would have the same energy whether it's bouncing back and forth or going in a straight line), so obviously the mechanism isn't that simple
23:41
< Rhamphoryncus>
Physics is hilarious: http://en.wikipedia.org/wiki/Temperature#Negative_temperature
23:43
<&McMartin>
"As the energy of the system increases in the population inversion, a system with a negative temperature is not colder than absolute zero, but rather it has a higher energy than at positive temperature, and may be said to be in fact hotter at negative temperatures."
23:51 Vash [Vash@Nightstar-e8057de2.wlfrct.sbcglobal.net] has joined #code
23:51 mode/#code [+o Vash] by ChanServ
23:53
< Rhamphoryncus>
McMartin: yeah, I read that as "the definition of temperature breaks in a rather spectacular fashion"
23:55
<&McMartin>
Thanks to the list below it and wikiwalking I have also learned the word "fulminologist"
23:55
<&McMartin>
A scientist that studies lightning.
23:56
<&McMartin>
Definitely the word of the day.
23:58
< Rhamphoryncus>
And unlike the crap that shows up in crossword puzzles and word searches it's actually a real word getting real usage!
23:58 You're now known as TheWatcher[T-2]
23:59
< Tarinaky>
TIL: A Thermonuclear Explosion is hotter than the core of the sun.
23:59
<&McMartin>
This also implies that when you "fulminate" about something that annoys you, you really are metaphorically shooting lightning bolts out your ears
--- Log closed Fri Aug 31 00:00:30 2012
code logs -> 2012 -> Thu, 30 Aug 2012< code.20120829.log - code.20120831.log >

[ Latest log file ]