←2009-06-12 2009-06-13 2009-06-14→ ↑2009 ↑all
00:01:34 -!- inurinternet has quit (No route to host).
00:14:13 -!- jix has quit (Read error: 113 (No route to host)).
00:29:34 -!- inurinternet has joined.
00:34:51 -!- nooga has quit (Remote closed the connection).
00:42:42 -!- Sgeo_ has joined.
00:52:30 <Asztal> did someone mention converting brainfuck into music recently?
00:53:13 <ehird> Asztal: brainfuck to fugue
00:53:14 <Asztal> I just found a brainfuck.mp3 buried away in my home directory :o
00:53:19 <ehird> and therefore that
00:58:20 -!- FireFly has quit ("Later").
01:00:11 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
01:17:40 <Asztal> http://tunes.org/~nef//logs/esoteric/07.05.19 <- apparently what I found was not actually generated from brainfuck :(
01:23:34 -!- Azstal has quit (".").
01:24:50 -!- psygnisfive has changed nick to augur_.
01:28:30 -!- augur_ has changed nick to augur.
01:37:10 -!- Sgeo has joined.
01:54:14 -!- Sgeo_ has quit (Read error: 110 (Connection timed out)).
01:57:15 -!- zzo38 has joined.
01:57:53 <zzo38> There is one thing you have forgotten in compile to BF
02:00:04 <zzo38> You forgot BrainClub
02:00:14 <zzo38> BrainClub compiles into brainfuck
02:03:38 <zzo38> asiekierka joins MegaZeux channel as well
02:05:22 -!- zzo38 has quit (Read error: 104 (Connection reset by peer)).
02:09:37 -!- amca has joined.
02:48:11 -!- MizardX- has joined.
02:48:17 -!- MizardX has quit (Read error: 54 (Connection reset by peer)).
02:48:44 -!- MizardX- has changed nick to MizardX.
02:59:10 -!- iEhird has joined.
02:59:23 <iEhird> why am I here?
02:59:37 -!- iEhird has quit (Remote closed the connection).
03:17:44 -!- Sgeo has quit (Read error: 104 (Connection reset by peer)).
03:17:46 -!- Sgeo_ has joined.
03:18:59 -!- oerjan has joined.
03:25:46 <augur> who knows
03:25:49 <augur> oh hes gone
03:25:55 <augur> nurrmind
03:26:04 <oerjan> maybe i know. what was the question anyhow? ;D
03:27:11 <oerjan> <pikhq> Man. We are growing organs *right now*.
03:27:22 <oerjan> this is the future. get used to it.
03:28:40 -!- puzzlet_ has joined.
03:30:48 -!- puzzlet has quit (Read error: 104 (Connection reset by peer)).
03:30:51 <oerjan> <ehird> ais523: I'm not sure it's a cellular automata, as things need to travel on paths, and cells can only affect cells that hit them (and often transform themselves in the process, or release new sparks)
03:31:38 <oerjan> if the state of a cell depends only on the previous state of a bounded neighborhood, then it's a cellular automaton. assuming it's all in a grid.
03:32:40 <oerjan> doesn't matter how complicated it is. you may still be out of look implementing it on a standard CA program though...
03:36:21 <oerjan> <ehird> is there a reversible, turing complete CA? <-- almost certainly
03:36:49 <oerjan> any reversible turing machine is a reversible CA, essentially
03:37:38 <oerjan> for the same reason stated 4 lines above
03:48:19 <Saites> I think BBM is a reversible Turing-complete CA.
03:48:30 <Saites> If you start with an infinite pattern.
03:49:09 <Saites> If you don't want to start with an infinite pattern, extend BBM so that you get a machine that grows.
03:49:33 <oerjan> what's BBM?
03:49:52 <Saites> Some CA. :-P
03:50:12 <Saites> Specifically, a Margolus CA that I think was described in "Cellular Automata Machines".
03:50:19 <Saites> It's available with MCell.
03:50:55 <oerjan> oh, and with a reversible CA you _cannot_ get a machine that grows, unless it grows towards the past as well
03:51:13 <oerjan> that might not be such a problem actually
03:51:41 <Saites> Can the CA be partial?
03:52:00 <oerjan> no
03:52:12 <Saites> Well, okay.
03:52:22 <Saites> Start with a partial CA that works. Find the holes in it and connect them randomly.
03:52:47 <oerjan> although "billiard ball machine"?
03:52:57 <oerjan> Saites: that might work
03:53:15 <oerjan> there is a slight problem that reversibility of CA's isn't entirely local
03:53:16 <Saites> The CA is a decent approximation to the actual billiard ball model.
03:53:21 <Saites> And that is what it stands for.
03:53:39 <Saites> The thing is, when balls collide, they also slow down.
03:53:59 -!- GreaseMonkey has joined.
03:54:01 <oerjan> we were recently discussing the Billiard Ball Machine on the wiki
03:54:16 <oerjan> (which is sort of an esolang)
03:54:41 <oerjan> hm there was no slowdown in that
03:56:13 <Saites> Indeed.
03:56:27 <Saites> In the CA, they slow down. In the actual billiard ball model, they don't.
03:56:46 <oerjan> ah
03:58:05 * oerjan thinks a link from the wiki is appropriate
04:00:56 -!- MizardX- has joined.
04:01:08 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
04:01:16 <oerjan> the link http://cell-auto.com/bbm/2d/index.html says it's universal too
04:01:26 -!- Sgeo has joined.
04:01:29 -!- MizardX- has changed nick to MizardX.
04:03:27 * oerjan finally remembers that Saites isn't a new guy :D
04:04:35 -!- Saites has changed nick to Petalon.
04:04:42 <Petalon> Muahaha! Now you have to remember all over again!
04:04:50 * oerjan swats Petalon mercilessly -----###
04:05:20 <Petalon> By the way, using this random Ancient Greek word generator, it took three tries to get something that isn't also an English word.
04:06:09 <oerjan> neologism is hard
04:06:17 <oerjan> also, a greek word
04:06:44 <Petalon> Neologism is a Greek word?
04:06:53 <Petalon> Neo and logos are Greek words, perhaps.
04:06:54 <oerjan> at least the parts are
04:07:09 <oerjan> probably needs a case suffix
04:07:27 * Petalon grabs more Greek words.
04:07:36 <pikhq> Saites = ?
04:08:01 <Petalon> Someone from Sais, or something.
04:08:13 <oerjan> english: saite
04:08:18 <pikhq> Petalon = ?
04:08:29 <oerjan> i pantaloni!
04:08:34 <Petalon> Leaf or slice.
04:09:08 <Petalon> Monas, Kolobathristes, Gramma, Teleo, Sumposion!
04:09:16 <Petalon> I think "sumposion" would make a fine English word.
04:09:25 <pikhq> DAMMIT IHOPE!
04:09:48 <oerjan> neologismos
04:09:49 <Petalon> Oh, I think it already is an English word. It's just "symposium".
04:10:13 <oerjan> hm wait is neo- correct greek, actually? it's ne_a_polis, after all
04:10:43 <Petalon> Monas, Kolobathristes, Gramma, Teleo, Paramenei!
04:10:55 <oerjan> teleo- is a fine english prefix
04:11:29 <oerjan> and gramma isn't particularly unknown either
04:11:43 <oerjan> is monas ~= monad, perhaps
04:12:03 <oerjan> now kolobathristes, i grant you.
04:14:07 <Petalon> Monas ~= monad indeed, though I don't know what ~= means.
04:14:19 <Petalon> You generally see it in English as mono-.
04:14:46 <oerjan> approximately equal
04:15:16 <oerjan> um i would assume monas is mono- + some suffix
04:15:45 <Petalon> A really tiny suffix, maybe.
04:17:33 <oerjan> like -r is a tiny suffix in "writer"?
04:17:52 <Petalon> Yeah.
04:17:54 <augur> keep in mind that in greek, there are certain phonological process that apply to s, t, and d in word final position
04:18:25 -!- Sgeo_ has quit (Read error: 110 (Connection timed out)).
04:18:31 <oerjan> augur: which is how i guessed it was monad in the first place
04:18:39 <augur> ey?
04:18:59 <oerjan> <Petalon> Monas, Kolobathristes, Gramma, Teleo, Paramenei!
04:19:40 * oerjan wishes crows could be exterminated
04:19:43 <augur> petalon: sumposion is written how, in original greek?
04:19:47 <oerjan> NOISY PESTS!
04:20:43 -!- pikhq has quit (Read error: 60 (Operation timed out)).
04:22:05 -!- pikhq has joined.
04:22:10 <oerjan> augur: Atlas is a particularly nice example, i think
04:22:20 <augur> of?
04:22:23 <oerjan> the s is actually a reduced -nt, it seems
04:22:26 <augur> petalon!
04:22:29 <augur> yes, it is.
04:22:48 <augur> but that only is true of some final s's
04:22:55 <augur> some words have underlying /s/
04:22:55 <oerjan> well, duh
04:23:17 <augur> i did this for a phonology test once, actually
04:23:30 <augur> well, we did greek for homework, and then a similar latin problem for the test
04:24:00 <oerjan> latin -s can be much the same, i recall
04:24:16 <augur> yes
04:24:26 <augur> it was an almost identical process
04:25:47 <oerjan> although sometimes the change seems to go the other way, with s -> r between vowels
04:26:31 <augur> flosis floris
04:26:40 <augur> thats a case of dissimilation, i think
04:26:53 <augur> i dont think its related to the whole d/t/s/nt thing
04:26:55 <oerjan> venus veneris
04:27:15 <augur> intervocalic dissimilation, yeah
04:35:49 -!- Sgeo_ has joined.
04:41:04 -!- oerjan has quit ("Reloxing").
04:50:41 <Petalon> Sumposion in the original Greek?
04:51:32 <Petalon> συμπόσιον
04:53:22 -!- Sgeo has quit (Read error: 110 (Connection timed out)).
05:00:34 <augur> ok
05:00:48 <augur> so symposium isnt too far off
05:01:07 <augur> the upsilon there should be transliterated as a y not a u
05:01:29 <augur> and since the word was loaned into latin before being brought into english, we get um instead of on
05:07:56 <GregorR> You know what sucks worse than a cluttered mess of a movie? Seeing it with friends who actually /enjoyed/ the crappy thing.
05:08:32 <augur> which movie
05:10:31 <Petalon> "Up".
05:11:37 <augur> aww, it was cluttered? :(
05:11:39 <augur> it looks so funny
05:12:11 <Petalon> Note that I am not GregorR.
05:12:51 <Petalon> I do not have the right to speak on his behalf, only to pretend to do so.
05:13:20 -!- MizardX- has joined.
05:13:21 * GregorR reappears.
05:13:22 <augur> i havent seen up
05:13:23 <GregorR> Pelham 123
05:13:24 <augur> but
05:13:26 <GregorR> It was really awful.
05:13:30 <augur> ahh how so?
05:13:37 <GregorR> Just a sec, talking on phone.
05:13:40 <augur> k
05:13:44 <augur> ill just watch it, dont worry
05:14:06 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
05:14:16 -!- MizardX- has changed nick to MizardX.
05:14:34 <GregorR> It was a cluttered mess. Particularly in terms of graphical effects.
05:14:42 -!- evenant has joined.
05:15:02 <GregorR> They'd use some effect for one scene, that's utterly unlike any other effect they'd used, and then *boom* right back into straight live action. And I'm thinking, "What the hell was I supposed to get from that effect? Other than 'confused'?"
05:16:00 <augur> hmm
05:16:01 <GregorR> And the main "hook" to the movie, what was supposed to make the villain interesting, was barely delved into at all, it was basically mentioned as a passing remark in two or three places. Oh, so that's why he's here. Well, back to the formula for hostage flicks then!
05:16:28 <GregorR> Other than that it was just utterly by-the-book. Down to at least four instances of "THREE - TWO - ONE" *saved by the bell*
05:17:04 <augur> mma watch it right now
05:22:34 -!- evenant has quit (Read error: 60 (Operation timed out)).
05:26:41 <augur> i cant find pelham 123 on the sites
05:27:35 <GregorR> It's too garbage to be ripped :P
05:28:33 <augur> probably
05:31:51 -!- nescience has quit (Read error: 113 (No route to host)).
05:43:13 -!- puzzlet_ has quit (Read error: 104 (Connection reset by peer)).
06:18:51 -!- Sgeo_ has quit ("Leaving").
06:40:55 -!- GuestShadowSkunk has joined.
06:54:43 -!- Slereah has quit (Read error: 110 (Connection timed out)).
07:33:02 <augur> cocks.
07:34:23 <amca> hens.
07:34:47 <augur> drakes.
07:35:45 <amca> stumped.
07:38:24 <augur> the female term is "duck"
07:38:42 <amca> Ah, poo.
07:38:46 <augur> or "hen", depending on species.
07:39:00 <augur> well, not depending on species sorry
07:39:03 <augur> but depending on person
07:44:55 -!- GreaseMonkey has quit ("YES -> thor-ainor.it <- THIS IS *DELICIOUS*!").
07:45:33 <GregorR> Masterpiecemachine has OpenID support now.
07:59:59 -!- clog has quit (ended).
08:00:00 -!- clog has joined.
08:01:08 -!- kar8nga has joined.
08:11:21 <Petalon> augur: may I suggest "dogs"?
08:11:31 <augur> what?
08:11:50 <Petalon> Well, you said "cocks" and "drakes".
08:12:00 <Petalon> Perhaps you might want to consider saying "dogs" as well.
08:12:12 <augur> that was years ago
08:12:27 <Petalon> Oh, you're right. Sorry.
08:24:34 -!- puzzlet has joined.
08:51:03 -!- MizardX has quit (Read error: 104 (Connection reset by peer)).
08:51:14 -!- MizardX has joined.
09:05:38 -!- kar8nga has quit (Remote closed the connection).
09:09:16 -!- sebbu2 has joined.
09:17:06 -!- nooga has joined.
09:24:55 -!- sebbu has quit (Read error: 110 (Connection timed out)).
09:37:49 <augur> http://www.youtube.com/watch?v=cCdbZqI1r7I
09:38:03 <augur> jeff hawkins doesnt understand why a universal turing machine is universal
10:06:41 <augur> tho maybe he was simplifying
10:06:53 <augur> we're from the same town in new york :D
10:10:56 <nooga> erm
10:18:41 -!- kar8nga has joined.
10:18:49 -!- MigoMipo has joined.
10:20:58 -!- MizardX has quit (Read error: 110 (Connection timed out)).
10:33:38 <augur> i want to build an HTM :O
10:33:40 <augur> it looks fun
10:33:51 <augur> we should build a language for an HTM system
10:34:59 -!- BeholdMyGlory has joined.
10:37:06 <amca> HTM? Whats that?
10:43:59 <lifthrasiir> hardware transactional memory?
11:01:05 -!- AnMaster has joined.
11:01:59 <augur> hierarchical temporal memory
11:02:06 <augur> he REALLY doesnt know what universal means in turings sense
11:02:13 <augur> he says brains arent universal in turings sense
11:02:16 <augur> which is obviously false
11:07:48 <amca> Im pretty sure brains dont develop infinite tapes
11:12:55 <AnMaster> hi
11:13:13 <AnMaster> ehird, as I said, you will see the pictures! ;P
11:13:15 <AnMaster> but
11:13:17 <AnMaster> not today
11:13:31 <AnMaster> I spent the last 21 hours on a train.
11:14:19 <AnMaster> sleeping wagon, but there were lots of issues, so the trip lasted about two hours longer than planned.
11:16:52 <AnMaster> First a broken switch just outside Kiruna, where the train had to wait for it to be repaired before it could continue. Then there were issues with the locomotive, they had to replace it somewhere north of Umeå... Then some other train that was in front of us and had issues.. and so on.
11:17:16 <AnMaster> wow... 360 new spam
11:17:19 -!- augur has quit (Read error: 54 (Connection reset by peer)).
11:17:54 -!- psygnisfive has joined.
11:19:14 -!- kar8nga has quit (Read error: 60 (Operation timed out)).
11:20:29 <AnMaster> fizzie, what was that program you used for panoramas?
11:22:37 <fizzie> AnMaster: hugin, which is mostly just an interface for different tools.
11:23:09 -!- MizardX has joined.
11:23:27 <AnMaster> thanks
11:23:34 <fizzie> It uses autopano-sift-c for automatic control points, nona for the combining, enblend/enfuse for blending and so on. I guess the camera-position-optimizer is in hugin itself, though.
11:24:14 <psygnisfive> amca, its true that brains dont have infinite tapes
11:24:30 <psygnisfive> but thats effectively irrelevant to the issue
11:25:02 <fizzie> (Also the UI is pretty idiosyncratic.)
11:25:04 <psygnisfive> a human WITH an infinite tape is a utm, and the only difference is memory size, nothing else
11:25:24 <psygnisfive> but then, your computer isnt a utm either for the same reason
11:25:52 <AnMaster> hm
11:26:18 <lifthrasiir> bounded storage machine, iirc.
11:26:34 <psygnisfive> sure. aka a finite state machine.
11:26:37 <AnMaster> fizzie, any pitfalls in particular?
11:26:45 <amca> I wonder what the latency and seek time on a universal turing machine tape is.
11:26:47 <AnMaster> fizzie, also, all pics I have are in raw format...
11:26:57 <psygnisfive> amca: what
11:27:23 <AnMaster> fizzie, or 16-bit (per channel) tiff using adobe rgb
11:27:38 <AnMaster> I guess I need to convert first
11:27:42 <amca> psygnisfive: I dont know what it is
11:27:48 <psygnisfive> what what is
11:28:39 <AnMaster> fizzie, there are packages for both autopano-sift and autopano-sift-c... what's the difference?
11:28:56 <fizzie> The -c version is the C port.
11:28:59 <amca> The latency and seek time on an infinite turing machine tape
11:29:05 <fizzie> From the .Net (probably C#) original.
11:29:07 <psygnisfive> ...
11:29:15 <AnMaster> fizzie, ah..
11:29:15 <fizzie> I think at least my version of hugin used autopano-sift-c by default.
11:29:16 <psygnisfive> there is no latency or seek?
11:29:34 <fizzie> There's some sort of patent issue on the SIFT algorithm, if you want to be wary of that sort of thing.
11:29:54 <AnMaster> fizzie, not in Europe I guess?
11:30:07 <lifthrasiir> psygnisfive: i wonder if there is a complexity class which computer solves efficiently but human brain doesn't.
11:30:19 <psygnisfive> sure, its called mathematics.
11:30:32 <psygnisfive> atleast ignoring savants
11:30:34 <fizzie> Anyway, the automagical control-point setting is not that important; I've been doing it manually just for the fun of it. Hugin auto-optimizes the points anyway, so you mostly have to set one point, then just randomly click on regions that are in both pictures of a pair and it'll figure it out. Assuming there's some sort of clear border/pattern there.
11:31:06 <AnMaster> fizzie, mostly forest and mountains (with snow on the tops.
11:31:08 <AnMaster> )
11:31:23 <AnMaster> anyway, this will take a while... got some other stuff to do first
11:31:30 <lifthrasiir> psygnisfive: if computational capacity of human brain is poorer than computer's (thus turing machine), there should be such problem... but i don't know any instance.
11:31:36 <lifthrasiir> such a problem*
11:31:45 <psygnisfive> what?
11:32:04 <psygnisfive> the computational class of both, ignoring memory size and looking just at the sort of operations on memory that can be performed, are identical.
11:32:15 <AnMaster> fizzie, it does handle tiff or other non-lossy formats right?
11:32:36 <lifthrasiir> psygnisfive: i think so, and i'm (mostly) answering to amca ;)
11:32:58 <psygnisfive> amca's question was about seek times
11:33:02 <psygnisfive> which makes no sense
11:33:06 <lifthrasiir> ah
11:33:35 <fizzie> AnMaster: I would presume so, yes. The output is generated as a TIFF, anyway. Though it's easier if you have still EXIF info in the source images, since it can figure out sensible lens parameters automatically.
11:33:57 <psygnisfive> a tape machine does not perform memory access like a von neumann machine does
11:34:25 <AnMaster> fizzie, my camera stores the exif separately. Like: 00004.mrw 00004.(something I forgot, which contains the exif)
11:34:28 <AnMaster> bbiab
11:34:29 <psygnisfive> you can simulate that, ofcourse, with the right state machine, but ultimately, thats a simulation, not a genuine aspect of the tape machine
11:35:04 <psygnisfive> further more, if we BUILT a tape machine with real tape and so forth, the speed of the machine is directly related to the speed of the state machine's components
11:35:11 <psygnisfive> and also to the motor that moves the tape around.
11:35:19 <amca> If a turing machine doesnt have latency or seek times, then time would be irrelevant to it, but time is a necessary component of state systems. There needs to be a time before and a time after state changes, or the concept of state doesnt make sense.
11:35:39 <psygnisfive> but seek and latency are concepts of memory access
11:35:46 <psygnisfive> tape machines do not perform memory access
11:36:12 <psygnisfive> all they do is move left or right one space and do something to the tape cell.
11:36:16 <fizzie> AnMaster: I'm not sure if the gui supports "use this image but exif from this", but it should be able to read EXIF data from TIFF images too, so if you convert your raw format to that, you could then use some of the command-line exif tools to stick the information there.
11:36:19 <psygnisfive> move, do something, move, do something
11:36:22 <psygnisfive> thats it.
11:36:30 <psygnisfive> there is no memory access on a tape machine
11:36:35 <psygnisfive> furthermore, lets take this seriously, ok
11:36:40 <psygnisfive> suppose we do have an infinite tape
11:36:47 <psygnisfive> if its finitely thing, it has infinite mass
11:36:49 <amca> The tape (and symbols on it) is the memory, and the tape moving left and right (as well as symbol reads/writes) are the access
11:36:56 <psygnisfive> and therefore the tape itself couldnt move so the device would have to
11:37:08 <psygnisfive> but if it were infinite in mass, the universe would collapse into a black hole
11:37:14 <lifthrasiir> amca: if you call that the "access", you're right. but it doesn't make any difference really.
11:37:33 <lifthrasiir> since it is computable anyway.
11:37:40 <psygnisfive> now if you say this is absurdly unrealistic and irrelevant to the question
11:37:40 <amca> hmmm
11:37:46 <psygnisfive> lets pretend we COULD do this magic with the tape
11:37:48 <psygnisfive> ok fine
11:37:58 <psygnisfive> lets also pretend we can do magic where the tape or the state machine moves infinitely fast
11:38:07 <psygnisfive> therefore theres no latency or seek time.
11:38:18 <psygnisfive> tape machines do not perform seek operations
11:38:28 <psygnisfive> they do not call out to l2 cache, ram, or to the harddrive
11:38:46 <psygnisfive> they just work directly with the infinite memory tape
11:38:53 <amca> In that case, wouldnt every symbol/part on the tape be in a supposition of states?
11:39:01 <psygnisfive> there is no memory access, just move-affect-move-affect
11:39:04 <lifthrasiir> amca: time is not relevalent, only thing relevalent is whether it's computable or not.
11:39:12 <psygnisfive> no why would it be
11:39:23 <lifthrasiir> relevant*
11:39:24 * amca would call moving left or right on the tape seeking.
11:39:30 <psygnisfive> amca
11:39:33 <psygnisfive> there is no seeking
11:39:36 <psygnisfive> its not LOOKING for something
11:39:44 <psygnisfive> it doesnt keep moving until it finds what its looking for
11:39:51 <psygnisfive> unless you write a program that simulates that behavior
11:40:10 * amca shrugs
11:40:14 <psygnisfive> EVERY step is a single move in one direction or another (or no move at all if your version of tape machines allows that)
11:41:05 <psygnisfive> a turing machine does not have latency or seek times because those concepts are not relevant here.
11:41:11 <amca> Isnt symbol change another step? Or do you mean step in a different sense?
11:41:23 <psygnisfive> symbol change is part of the transition
11:41:32 <psygnisfive> "write A then move right"
11:41:36 <psygnisfive> "write B then move left"
11:41:39 <psygnisfive> etc.
11:41:51 <amca> Arent they both two steps each?
11:41:53 <psygnisfive> those are the actions taken o transitions.
11:42:05 <psygnisfive> well, it they're an infinite number of steps if you want it to be
11:42:26 <psygnisfive> "move write head down, hold for 10 seconds, life write head"
11:42:46 <psygnisfive> "engage write head motor, initiate power on, do this do that"
11:43:11 <psygnisfive> and suppose we define "seek time" as the time between read cycles on the machine
11:43:20 <psygnisfive> ok, now we're just dealing with an engineering question
11:43:30 <psygnisfive> what is the seek time on some particular device we've constructed
11:43:34 <psygnisfive> i dont know, go build one
11:43:50 <psygnisfive> double the voltage and you could probably double the speed
11:43:57 <psygnisfive> so its a pointless question
11:44:26 <psygnisfive> its entirely dependent on a) a pointless definition of "seek time" on a machine that has no similar concept to seeking
11:44:42 <psygnisfive> and on b) the particulars of how you construct the physical device
11:44:52 <psygnisfive> its like asking whats the seek time on a hard drive
11:44:56 <psygnisfive> well, what harddrive?
11:44:59 <psygnisfive> the one in my laptop?
11:45:12 <psygnisfive> the refrigerator sized 40 meg harddrives of the 70s?
11:45:30 <psygnisfive> a solid state hard drive in a macbook air?
11:45:57 <psygnisfive> such a question about turing machines is meaningless
11:46:03 <psygnisfive> its like asking how many rpms my monitor gets
11:46:07 <psygnisfive> good night.
11:47:27 <lifthrasiir> amca: turing machine is an abstract concept, not governed by physical laws. so "time" is not relevant (though certain problems have well-defined time characteristics, for example, "within a polynomial factor of problem size") and whether it will be able to solve or not is relevant.
11:48:40 * amca doesnt understand how you can have "state" without "time"
11:50:16 <lifthrasiir> amca: time could be relevant for practical purpose, of course. but as long as it takes finite time, it is equivalent to turing machine (and vice versa).
11:51:12 <lifthrasiir> there are several examples of problems which cannot be solved by turing machine, and in turn computers at all.
11:51:30 <lifthrasiir> in turn by computers at all*
11:54:21 -!- inurinternet has quit (Connection timed out).
11:56:27 <AnMaster> <fizzie> AnMaster: I'm not sure if the gui supports "use this image but exif from this", but it should be able to read EXIF data from TIFF images too, so if you convert your raw format to that, you could then use some of the command-line exif tools to stick the information there.
11:56:32 <AnMaster> well, I can do that
11:56:39 <AnMaster> but no idea how to copy the exif
11:56:41 <AnMaster> to that tiff
11:56:56 <AnMaster> I usually use gimp + ufraw to convert my raw images
11:57:04 <AnMaster> never tried to do anything about the exif before
11:57:33 <fizzie> "exiftool -TagsFromFile file_with_exif.ext generated_tiff.tif" should probably copy it.
11:57:50 <AnMaster> fizzie, exiftool? what package
11:57:55 <AnMaster> I have no such thing here
11:58:09 <AnMaster> oh wait... media-libs/exiftool
12:10:32 -!- jix has joined.
12:10:33 -!- KingOfKarlsruhe has joined.
12:11:44 -!- FireFly has joined.
12:29:29 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
12:29:48 -!- MigoMipo has joined.
12:30:53 -!- MigoMipo has quit (Client Quit).
12:31:11 -!- MigoMipo has joined.
12:31:43 -!- kar8nga has joined.
12:41:27 -!- MizardX- has joined.
12:46:53 -!- MizardX has quit (Read error: 60 (Operation timed out)).
12:47:23 -!- MizardX- has changed nick to MizardX.
12:56:12 -!- Corun has joined.
13:19:55 <AnMaster> fizzie, did you do that "lens calibration" thingy before?
13:20:00 -!- asiekierka has joined.
13:20:01 <asiekierka> Hi
13:20:23 <fizzie> AnMaster: No, didn't bother.
13:20:27 <asiekierka> Heh, I implemented BF on a game creation system
13:20:31 <AnMaster> fizzie, ah
13:20:35 <asiekierka> result: Why did I bother, it lags as hell of all hells
13:21:11 <asiekierka> But I implemented something! :DD
13:21:20 <AnMaster> asiekierka, Some years ago I did game of life in a rpg game engine for Mac OS Classic (so won't run on Intel Macs iirc)
13:21:25 <amca> asiekierka: Which game creation system?
13:24:02 <asiekierka> amca: MegaZeux
13:24:11 <AnMaster> fizzie, it doesn't seem to be able to handle tiff...
13:24:13 <AnMaster> very strange
13:24:16 <asiekierka> it's multiplatform
13:25:50 <amca> Is that for making text/console based games?
13:25:55 <fizzie> AnMaster: Hm. Strange. It *should*. The panotools 16-bit workflow guide -- at http://wiki.panotools.org/16bit_workflow_with_hugin -- says "The TIFF images can be loaded into hugin as per usual except --" (and the exception is about the EXIF thing).
13:26:16 <AnMaster> fizzie, what version of hugin
13:26:22 <AnMaster> I have 0.7.0-r1 here now
13:26:29 <AnMaster> maybe I need 0.8.0-rc3?
13:26:51 <asiekierka> amca: Yeah
13:26:56 <fizzie> AnMaster: Possibly. I have 0.7.0 too, but on the other hand I haven't tried to use non-jpeg images either.
13:26:57 <asiekierka> Don't confuse with ZZT, MegaZeux is much more advanced
13:27:04 <AnMaster> ah
13:27:12 <asiekierka> If we compare them to operating systems, ZZT is DOS and MegaZeux is Linux
13:27:45 <amca> Did ZZT inspire it?
13:27:49 <asiekierka> Sure!
13:28:00 <asiekierka> ZZT was what started the text/console GCS's
13:28:03 <asiekierka> made by Tim Sweeney
13:28:08 <fizzie> I haven't bothered to get more than 3*2 gigabytes of storage space, and had trouble even with those three-megabyte .jpgs; the raw files are a bit too big for "let's take some 1300 snaps" vacation photograpy.
13:28:11 <asiekierka> AFAIK it was coded in Pascal but I might be wrong...
13:28:26 <asiekierka> MegaZeux was coded in C/ASM and the multiplatform port was ported to C++/SDL and later on to C/SDL
13:28:34 <asiekierka> so it can run on many platforms
13:28:49 <asiekierka> the DS (no audio), the PSP, the Wii, the PS3, Mac, Linux, Windows...
13:28:54 <asiekierka> the GP2X
13:29:02 <asiekierka> the newer Amigas
13:29:06 <asiekierka> (er, i mean, AmigaOS)
13:29:43 <asiekierka> MegaZeux 2.70 and lower can run only on DOS, while the port (2.80 and higher) can run on nearly anything with C and SDL. No-one cares about DOS though.
13:29:46 <amca> asiekierka: U thought of porting it to BF and then running your MegaZeux BF on it?
13:29:58 <asiekierka> amca: There's no SDL for BF
13:30:07 <asiekierka> but I could port it to CrainF**k if I was way too bored
13:30:14 * amca still cares for poor, neglected DOS
13:30:18 * asiekierka too
13:30:25 <asiekierka> I even wanted to add a native BF interpreter to MegaZeux
13:30:29 <AnMaster> <fizzie> I haven't bothered to get more than 3*2 gigabytes of storage space, and had trouble even with those three-megabyte .jpgs; the raw files are a bit too big for "let's take some 1300 snaps" vacation photograpy.
13:30:31 <AnMaster> ouch
13:30:32 <AnMaster> indeed
13:30:37 <AnMaster> I have around 300 photos
13:30:53 <AnMaster> I bought a 4 GB memory card while up there in Kiruna
13:31:32 <asiekierka> I don't get how to make logic stuff (AND, OR, XOR) run in BF
13:31:34 <asiekierka> cuz NOT, i know
13:31:41 <asiekierka> the rest, I know NOT
13:31:44 <asiekierka> er i mean
13:31:45 <asiekierka> i don't know
13:32:26 <AnMaster> ..
13:32:36 <AnMaster> asiekierka, what do you mean
13:33:05 <fizzie> Yes, well, in fact I only had 2*2 GB with me, and had to buy a horribly overpriced 2 GB card from Sorrento, since I couldn't be bothered to leave the tourist-infested areas in search of a reasonable electronics/computer-stuff shop.
13:33:51 <AnMaster> fizzie, ah... I went to one of the normal stores there. Since that town is not very tourist-y yet. That starts in 1-2 weeks
13:34:08 <AnMaster> I always prefer coming either before or after the most tourist intensive periods
13:36:15 <fizzie> We were a bit out-season too, but not enough. And Sorrento's really a tourist-only place. Oh well, the card was just three times the price it would've been in the local supermarket here at home. :p
13:36:50 <AnMaster> wth...
13:37:08 <AnMaster> my camera decided to overlap some of the image filenames
13:37:11 <AnMaster> between the cards
13:37:22 <AnMaster> it is set to give them ever-increasing sequential numbers
13:37:28 <AnMaster> except here it overlapped 5 or so
13:37:32 <AnMaster> near the middle
13:37:41 <AnMaster> so would be a PITA to correct.
13:38:44 <fizzie> (Off to go buy some food-supplies.)
13:39:00 <AnMaster> cya
13:42:02 <AnMaster> I hate USB 1.1
13:46:51 -!- KingOfKarlsruhe has quit (Remote closed the connection).
13:50:50 <asiekierka> I want to make an esolang which has both digital and analog components
13:54:07 -!- Corun has quit (Read error: 110 (Connection timed out)).
13:54:28 <ehird> <reddit> Analyst: Dell should buy Palm
13:54:32 <ehird> how random...
13:54:35 <nooga> http://www.opera.com/freedom/ what now? :F
13:54:40 <ehird> hiamcahaven'tseenyouaroundETC
13:55:01 <ehird> nooga: yah opera employees on reddit have been shifty about that
13:55:15 <ehird> is probably bullshit
13:55:19 <ehird> amca: what brings you here
13:55:21 <nooga> i bet it's something stupid
13:55:37 <nooga> like new opera turbo
13:55:56 <nooga> or some kind of cloud contraption
13:56:14 <ehird> 02:36 oerjan: <ehird> is there a reversible, turing complete CA? <-- almost certainly
13:56:14 <ehird> i mean
13:56:16 <ehird> that has been described
13:56:17 <ehird> and used
13:56:21 <amca> ehird: I have an interest in programming languages and particularly esoteric ones
13:56:30 <ehird> 02:49 oerjan: what's BBM?
13:56:33 <ehird> billiard ball machine
13:56:39 <ehird> 02:50 oerjan: oh, and with a reversible CA you _cannot_ get a machine that grows, unless it grows towards the past as well
13:56:43 <ehird> erm growing is reversible, no?
13:56:49 <nooga> busy beaver? :>
13:57:05 <ehird> Saites = kerlo
13:57:06 <ehird> so he knows you were discussing bbm
13:57:06 <ehird> :)
13:57:54 <nooga> busy beavers are class of turing machines that run long sequences without going into cyclesAFAIR
13:57:55 -!- sebbu has joined.
13:58:07 <amca> ehird: Im also designing an esolang, so Im hoping to get help here when I get stuck
13:58:20 <nooga> amca: show us :F
13:58:30 <ehird> 08:37 augur: http://www.youtube.com/watch?v=cCdbZqI1r7I
13:58:31 <ehird> 08:38 augur: jeff hawkins doesnt understand why a universal turing machine is universal
13:58:40 <ehird> because irreducibly complex flagella
13:58:47 <amca> nooga: Show you what?
13:58:50 <ehird> 10:02 augur: he says brains arent universal in turings sense
13:58:52 <ehird> 10:02 augur: which is obviously false
13:58:52 <ehird> ehm
13:58:54 <ehird> my brain doesn't have infinite tape
13:59:06 <ehird> and I cannot do long computations without the aid of extra memory eg paper
13:59:11 <nooga> amca: your dick :C
13:59:18 <nooga> amca: your esolang ofc
13:59:23 <ehird> consciousnesses are not even FSM imo
13:59:24 <ehird> maybe brains are
13:59:29 <ehird> but not our conscious thought process
13:59:56 <nooga> ehird: that'd mean you can build conscious machine
13:59:59 <amca> nooga: It is only in design stage so I have nothing to show presently
14:00:21 <nooga> maybe you could describe it briefly, i'm curious
14:00:56 <ehird> 12:21 AnMaster: asiekierka, Some years ago I did game of life in a rpg game engine for Mac OS Classic (so won't run on Intel Macs iirc)
14:00:57 <ehird> emulator
14:01:49 <AnMaster> ehird, sure. But also it was some shareware one iirc.
14:01:59 <AnMaster> yes, it was before I went FOSS
14:02:01 <nooga> i thought that leopard can run binaries from PPC
14:02:13 <ehird> yes
14:02:15 <ehird> Rosetta
14:02:18 <ehird> but no more Classic mode
14:02:25 <ehird> but you could use a real full-hardware emulator
14:02:27 <ehird> 13:59 nooga: ehird: that'd mean you can build conscious machine ← wat?
14:02:41 <ehird> 13:55 nooga: or some kind of cloud contraption ← obvious from the logo
14:02:45 <ehird> 13:56 amca: ehird: I have an interest in programming languages and particularly esoteric ones ← hi
14:02:49 <ehird> 13:57 nooga: busy beavers are class of turing machines that run long sequences without going into cyclesAFAIR ← yes
14:02:52 <nooga> ehird: consciousnesses are not even FSM imo
14:02:52 <nooga> ehird: maybe brains are
14:03:12 <nooga> we can build FSMs
14:03:14 <ehird> nooga: "that'd mean you can build conscious machine", what is the issue with this?
14:03:21 <ehird> i completely think strong ai is possible
14:03:33 <ehird> we are just wetware running on crappy analog carbon computers
14:03:35 <amca> nooga: It is called minpostlisp, which is short for Minimal Postfix List Processor. It is a stack machine which only has conses and nils as data
14:04:01 <nooga> oh
14:04:15 <ehird> GregorR: that film has john travolta. it must suck by definition
14:04:34 <nooga> so it's like: reverse program string, interpret as lambda calculus
14:04:35 <nooga> ? :>
14:05:11 <amca> nooga: Nope. It is concatenative rather than applicative.
14:05:42 <nooga> any syntaxx examples yet?
14:05:44 <ehird> http://www.favbrowser.com/opera-freedom-mozilla-jetpack/
14:05:45 <ehird> Although it’s too early to speculate, especially when there is so little information revealed so far, we can still check few things while waiting. Lets begin with the logo.
14:05:47 <ehird> Oh well, maybe they just like clouds.
14:05:49 <ehird> hahahahahaha
14:06:03 <ehird> how can this moron say such things and not even know about the cloud computing hype
14:06:05 <ehird> what a dolt :)
14:06:11 <ehird> ANYWAY
14:06:17 <nooga> bollocks
14:06:24 <ehird> amca: sounds ... well, ok, not interesting, but more unique than most stuff :P
14:06:33 <ehird> check out underload for a concatenative tarpit.
14:07:00 <ehird> [[Opera Turbo alone is groundbreaking for most of the world. Why are you quoting the notorious and obsessive Opera troll “nobody”? Why are you always 100% biased against Opera?]]
14:07:09 <ehird> Opera fanboys, on the other hand, are eternally amusing.
14:07:42 <amca> nooga: They wont make any sense, as everthing is represented as conses and nulls. But, for example, the number one might be represented by ##-, where # represents nil, and - represents cons
14:07:42 <nooga> opera turbo is crap
14:07:57 <ehird> nooga: opera turbo is ok for things like shitty hotel/airport internet and mobile internet.
14:08:05 <nooga> i can't imagine who would use a browser that mangles all images for 0.5s less loading time
14:08:08 <ehird> but it's a horrendous violation of privacy which they don't make clear, and the images look like shit
14:08:16 <ehird> nooga: 0.5 if you have an ok internet connection
14:08:27 <ehird> mobile 3g = <50kb/s
14:08:29 <nooga> and the privacy
14:08:36 <ehird> airport/hotel wifi = who the fuck knows
14:08:45 <ehird> nooga: yah, that's bullshit, but your isp sees all your shit anyway
14:08:55 <ehird> and i can't imagine doing much sensitive while using mobile internet
14:08:57 <ehird> or w/e
14:09:09 <nooga> i dl some weird porn and chat about weed on IM ;F
14:09:15 <ehird> it's crap but less crap for its intended use case.
14:09:25 <amca> ehird: underload?
14:09:30 <ehird> amca: (note: me and nooga aren't exactly the usual sort of crowd here; this place is normally not this off-topic :p)
14:09:36 <ehird> amca: http://esolangs.org/wiki/Underload
14:09:43 <ehird> our wiki has everything! as long as oklopol didn't make it
14:11:23 <amca> ehird: I like chat tangents :) I refer to the esolang wiki all the time for ideas etc :)
14:11:27 <ehird> http://hex.alife.co.uk/diffusion/ ;; margolus, SWAP-on-DIAG: looks very demoscene
14:11:36 <amca> eep. Sorry for excessive smilies :)
14:11:38 <ehird> very pretty cellular automata
14:11:46 <ehird> amca: it's ok i'll cancel them out :( :( :(
14:12:01 <amca> heh
14:13:59 -!- sebbu2 has quit (Read error: 110 (Connection timed out)).
14:17:38 <nooga> i played with CA some time
14:17:46 <nooga> for*\
14:23:59 <ehird> ooh
14:24:04 <ehird> can you make BCT reversible, I wonder?
14:24:12 <ehird> by marking the ip
14:24:35 <ehird> bct is essentially a 1d ca
14:25:44 -!- kar8nga has quit (Remote closed the connection).
14:25:51 <ehird> i mean
14:26:00 <ehird> prog|data
14:26:01 <ehird> in prog,
14:26:26 <ehird> *0→in data, change *a into a*
14:27:04 <ehird> *1→x=next command; in data, if *1 instead of *0, add x to the end of the data
14:27:09 <ehird> in each case, change *x in prog to x*
14:27:15 <ehird> so
14:27:20 <ehird> *00111|*101
14:27:21 <ehird> goes:
14:27:36 <ehird> *00111|*101
14:27:44 <ehird> 0*0111|1*01
14:27:57 <ehird> 00*111|10*1
14:28:12 <ehird> 001*11|10*11
14:28:18 <ehird> etc
14:28:22 <ehird> the issue here is cycling
14:28:28 <ehird> but I'm sure you could fix that
14:28:33 <ehird> it's basically bct except you never throw anything away.
14:28:47 -!- nooga_ has joined.
14:28:53 <ehird> nooga_: hi
14:29:05 <ehird> did you see my stuff about reversible bct?
14:29:06 <nooga_> hi :F
14:29:10 <nooga_> nope
14:29:14 -!- nooga has quit (Read error: 60 (Operation timed out)).
14:29:19 <ehird> oh, wait, cycling isn't a problem, *a|b's previousness is a*|b
14:29:32 <nooga_> my connection is flaky
14:29:37 <ehird> nooga_: here
14:29:38 <ehird> :
14:29:50 <ehird> http://tunes.org/~nef/logs/esoteric/09.06.13
14:29:52 <nooga_> ?
14:29:59 <ehird> nooga_: go to the bottom, start from "can you make BCT reversible, I wonder?"
14:30:15 <ehird> nooga_: "→" is →
14:30:18 <ehird> clog messes the logs
14:31:02 <nooga_> uhm
14:32:10 <AnMaster> fizzie, about idiosyncratic GUIs... I agree Hugin has one of those. But it isn't much worse than that of ufraw. Took a while to learn ufraw. And I still have to check docs sometimes.
14:33:19 <nooga_> ehird: let me think a while
14:33:31 <nooga_> for*
14:33:34 <AnMaster> ehird, I may soon be ready to show you the first picture...
14:33:44 <ehird> AnMaster: Y'know, I'm not obligated.
14:33:50 <AnMaster> whoa 36.9 MB tiff
14:33:54 <AnMaster> :/
14:34:02 <AnMaster> it is 16 bits per channel though
14:34:51 <ehird> AnMaster: Is there any compression whatsoever on the images?
14:35:03 <AnMaster> ehird, yes, deflate was used
14:35:06 <AnMaster> tiff supports that
14:35:13 <ehird> AnMaster: non-lossless
14:35:16 <ehird> er, lossy that is
14:35:20 <ehird> from taking of picture to delivering
14:35:22 <AnMaster> ehird, deflate isn't lossy
14:35:33 <AnMaster> and no, so far there has been no lossy compression
14:35:35 <ehird> AnMaster: Yes, but any lossiness at any other point?
14:35:39 <ehird> yay
14:35:54 <ehird> i might just look then
14:35:57 <nooga_> what images?
14:36:02 <AnMaster> ehird, not yet. I might have to turn it into a jpg to be able to upload it. But that would be after all post-processing
14:36:09 <AnMaster> like, this is a bit too dark
14:36:14 * AnMaster goes to adjust curves
14:36:21 <ehird> AnMaster: KEEP THE ORIGINAL
14:36:26 <AnMaster> ehird, also I'm going to panorama-ise this
14:36:32 <ehird> nooga_: he took some sort of images of his holiday and i'm making it as hard as possible for him to show me them.
14:36:33 <AnMaster> so I need the high quality tiff
14:36:44 <AnMaster> ehird, yes I'm going to save the mrws
14:36:47 <AnMaster> the raw files
14:36:50 <AnMaster> they are smaller
14:36:53 <AnMaster> around 11 MB each
14:37:01 <AnMaster> I think it is 12 bits per channel there
14:37:05 <ehird> i just want gigabytes of stuff
14:37:09 <ehird> i like downloading gigabytes
14:37:14 <ehird> for archival purposes.
14:37:49 <AnMaster> ehird, heh
14:38:10 <ehird> there's just something about using terabytes of disk space that's fun.
14:38:47 <ehird> hmm i have something like 838GB of disk space around here
14:38:52 <AnMaster> ehird, it is wide-gamut atm.... (camera was set to use AdobeRGB, not sRGB)... So I need to use 16 bits all the way
14:38:56 <ehird> mac 250GB, old pc 80+500GB, iphone 8GB
14:39:15 <ehird> well some in my 2004 ipod too i guess but i'm not gonna count everything
14:39:25 <ehird> heck my router has some flash memory :P
14:40:20 -!- amca has quit ("Farewell").
14:42:16 <nooga_> today some brit stopped me and asked for a way
14:42:46 <AnMaster> I think the English term is "asking for directions", but I may be wrong.
14:42:58 <nooga_> ah yes
14:43:19 <nooga_> and i said "no understands no no"
14:43:21 <nooga_> D:D:D
14:43:48 <ehird> nooga_: YER DIRTY IMMIGRANT GET OUT OF THIS COUNTRY BRITISH NATIONAL PARTY 4EVA
14:44:53 <nooga_> i'm sitting in the capital of my cuntry and not going anywhere, especially THAT ISLAND
14:44:57 <nooga_> :F
14:45:00 <ehird> internet is sooo slow
14:45:07 <ehird> nooga_: Stop this cuntry!
14:49:10 <ehird> Anyway.
14:51:26 -!- Corun has joined.
14:59:36 <nooga_> :F
15:05:25 <pikhq> ehird: Man, the BNP and the UKIP.
15:05:46 <pikhq> "I HATE CURRY!" and "I HATE THE CONTINENT!"
15:06:04 <ehird> pikhq: UKIP's adverts have said things like "We are the only party supporting leaving the EU" and "Although our racist ideas are great, don't vote for extremist, racist parties, please."
15:06:10 <ehird> (My reaction: OK, I won't vote for UKIP then.)
15:06:22 <ehird> They're really desperate to avoid association with BNP.
15:06:32 <ehird> I mean, seriously. They're the only ones wanting to leave the EU? That's just a blatant lie.
15:06:35 <AnMaster> here we go...
15:06:37 <AnMaster> ufraw-batch --out-depth=16 --out-type=tiff --exif --zip --exposure=auto --wavelet-denoising-threshold=30 --temperature=5267 --green=1.023 --out-path=pano1 --create-id=also pict12{16..21}.mrw
15:06:41 * AnMaster hopes it works this time
15:06:48 <AnMaster> last time white balance just messed up
15:07:02 <pikhq> ehird: Never mind that leaving the EU is extremely dumb.
15:07:16 <ehird> yeah
15:07:26 <ehird> oh well
15:07:31 <ehird> the right will be stupid.
15:07:40 <nooga_> hm well
15:07:56 <pikhq> I'm kinda impressed that your right-wing nutjobs make the US's seem sane.
15:08:04 <nooga_> i'm not sure if EU is okay, i mean yeah it's cool to be part of the union
15:08:22 <nooga_> but sometimes i feel that EU might bring something bad in future
15:08:56 <nooga_> epecially that Polish government is a bunch of pussies
15:09:12 <asiekierka> ...
15:09:15 <ehird> the eu needs some reforms and shit buttt.
15:09:19 <ehird> pikhq: what? no
15:09:22 <asiekierka> I should feel offended, nooga_
15:09:26 <asiekierka> but in fact, I agree with you
15:09:27 <ehird> pikhq: the US is the most right-wing country in the world
15:09:32 <ehird> asiekierka: he's polish himself you dolt.
15:09:37 <asiekierka> ...
15:09:40 <asiekierka> he is?
15:09:51 <nooga_> no jestem przecież :D
15:09:55 <pikhq> ehird: That's because we've got right-wing and center.
15:10:07 <ehird> pikhq: What? Even your Democrats are right-wing.
15:10:26 <ehird> Seriously, look at a scandinavian country and say you're center :p
15:10:33 <asiekierka> I wonder which esolang didn't have a game made in it
15:10:42 <asiekierka> is there a game for Befunge?
15:10:43 * pikhq was thinking of the smaller parties, like the Greens. :P
15:10:48 <ehird> asiekierka: Most.
15:10:50 <ehird> And yes.
15:10:52 <ehird> Life, wumpus, ...
15:10:58 <asiekierka> ...In Befunge?
15:11:00 <ehird> pikhq: OK, the greens are center.
15:11:06 <ehird> asiekierka: Sure. One of the first programs.
15:11:08 <ehird> -93 that is.
15:11:09 <nooga_> HUNT TEH WUMPUS
15:11:12 <asiekierka> Hmm...
15:11:18 <ehird> well, I don't think they depend on space behaviour
15:11:18 <pikhq> ^help
15:11:19 <fungot> ^<lang> <code>; ^def <command> <lang> <code>; ^show [command]; lang=bf/ul, code=text/str:N; ^str 0-9 get/set/add [text]; ^style [style]; ^bool
15:11:20 <ehird> so I guess -98 too
15:11:30 <pikhq> Note: fungot is written in Befunge.
15:11:31 <fungot> pikhq: omfg my eyes cant believe what i think,no offence he's good! i was your vampire" performed by marilyn manson.its a instrumental but even democrats have a lying, and love the video, and i'll say again, quite unbelievable that you're arguing that this video
15:11:33 <asiekierka> What about Underloa---wait, you can't make a game in Underload
15:11:37 <ehird> !befunge http://catseye.tc/projects/befunge93/eg/wumpus.bf
15:11:38 <EgoBot> Screwed up making caves. Trying again.
15:11:39 <ehird> asiekierka: Yeh.
15:12:04 <asiekierka> I should make a game in Underload!
15:12:05 <AnMaster> fizzie,
15:12:07 <AnMaster> Filename /mnt/oldgentoo/photos/2009_jun_8-13_kiruna/10290611/pano1/pict1216.tif
15:12:07 <AnMaster> width 3272 height 2456
15:12:07 <AnMaster> reduce size to 1600 x 1201
15:12:10 <AnMaster> it just said that
15:12:12 <asiekierka> And I would if not for that it doesn't have any input
15:12:16 <AnMaster> any clue why it want to reduce the size
15:12:33 <ehird> AnMaster: it's an evil fucker that likes destroying information.
15:12:33 <pikhq> ehird: Anyways, I'm not saying we don't have more right-wingers, I'm saying your right-wing nutjobs are more right-wing than ours. ... Which is kinda scary.
15:12:47 <AnMaster> ehird, I'm trying to create a panorama...
15:12:54 <AnMaster> I just hope it will work :)
15:12:58 <ehird> pikhq: Ehm.. The BNP and UKIP say things that are considered socialist in the US.
15:13:10 <ehird> National healthcare, etc.
15:13:21 <pikhq> ehird: Oh, so they're national socialists. \o/
15:13:21 <myndzi> |
15:13:21 <myndzi> /`\
15:13:21 <pikhq> :P
15:13:35 <AnMaster> that's still one off
15:13:36 <nooga_> nazi skinheads O_o
15:13:36 <nooga_> ?
15:13:40 <ehird> pikhq: Shut up, nazism has nothing to do with socialism :P
15:13:51 <pikhq> ehird: Does in name.
15:13:51 <ehird> nooga_: No, not skinheads. But BNP's founder, eh, "Mein Kampf is my Bible."
15:13:53 <pikhq> ... But only in name.
15:13:58 <ehird> pikhq: xactly.
15:14:20 <ehird> hay letz vote for objectivist parties
15:14:21 <pikhq> nooga_: They're in favor of deporting non-English and stopping immigration.
15:14:39 <nooga_> well
15:14:42 <ehird> YO DAWG SO I HERD U LIEK FIGHTING YO DAWG SO I GAVE YOU CAPITALISM
15:14:55 <ehird> SO YOU CAN FIGHT YO DAWG WHILE YOU MAKE MONETARY TRANSACTIONS
15:15:15 <nooga_> still, tom green owns xzibit
15:16:04 <asiekierka> Did anyone make a game in Piet?
15:16:42 <ehird> yanky telefone
15:17:13 <nooga_> oh shiii
15:17:32 <nooga_> my train departs at 05:45
15:17:55 <nooga_> in the morning :E
15:18:12 <ehird> nooga_: ouch
15:18:40 <ehird> well one day i slept at 3am and got up 6am and had to be on a train an hour or two later
15:18:44 <ehird> but that's not quite as bad
15:19:06 <nooga_> you can still sleep in the train
15:19:16 <ehird> couldn't
15:19:23 <ehird> nooga_: what time is it there?
15:19:33 <nooga_> 16:19
15:19:45 <ehird> mm
15:23:17 <asiekierka> I'm looking for a good esolang to make a game in
15:23:21 <asiekierka> where no-one has made on
15:23:21 <asiekierka> e
15:23:26 <nooga_> SADOL :D
15:23:35 <nooga_> it should be trivial
15:23:45 <asiekierka> well
15:23:48 <asiekierka> the point is to not be trivial
15:23:53 <nooga_> bwh
15:23:54 <asiekierka> like making a game in Brainf**k
15:23:57 <nooga_> beh*
15:24:24 <asiekierka> also deriatives of BF don't count
15:24:30 <asiekierka> nor does Malbolge
15:24:34 <asiekierka> or any language without input
15:25:48 <nooga_> intercal?
15:26:23 <asiekierka> ...
15:26:39 <ehird> nooga_: been done
15:26:54 <asiekierka> oh
15:26:55 <ehird> http://www.muppetlabs.com/~breadbox/intercal/ins/description.html
15:27:00 <ehird> as a cgi to boot
15:27:03 <asiekierka> and any language with a game made in it already doesn't count
15:27:06 <nooga_> RUBE?
15:27:21 <ehird> asiekierka: you are so arbirary
15:27:24 <ehird> arbitrary
15:27:45 <Gracenotes> of course it's impossible to determine whether a game has been made in a language by anyone...
15:27:51 <asiekierka> well
15:27:55 <asiekierka> i only count the ones published
15:28:09 <asiekierka> the ones unpublished count NOT
15:28:22 <Gracenotes> and does, for example, "guess the number between 1 and 100" count as a game?
15:28:22 <asiekierka> ...I wonder why didn't any of you propose DOBELA xDDD
15:28:31 <Gracenotes> it's a common programming exercise..
15:28:31 <asiekierka> Gracenotes: See "Taking Over The World".
15:28:34 <pikhq> Note that your definition of trivial seems to include all Turing-complete languages with input and output.
15:28:36 <asiekierka> It's a game that does exactly that.
15:28:51 -!- M0ny has joined.
15:28:52 <Gracenotes> I'm not talking about a "derivative", as you've said
15:28:57 <nooga_> whitespace?
15:29:03 <asiekierka> pikhq: By "trivial" I mean "way easier than BF"
15:29:18 <asiekierka> nooga_: checking whitespace
15:29:25 <nooga_> :@
15:29:44 <nooga_> probably you'll need syntax highlighting
15:29:55 <asiekierka> probably I will make a converter
15:29:57 * Gracenotes notes his question was avoided
15:30:01 <asiekierka> i first input abc
15:30:06 <nooga_> asiekierka: that does not count
15:30:13 <asiekierka> then I replace a with space, b with tab and c with lf
15:30:22 <asiekierka> nooga_: yes it does
15:30:30 <nooga_> that will count as cheating
15:30:41 <asiekierka> I can write a syntax highlighting app that shows a space as a
15:30:43 <asiekierka> a tab as b
15:30:46 <asiekierka> and an lf as c
15:31:18 <Gracenotes> it might be even more useful to show combinations of tabs and spaces as various characters
15:31:29 <asiekierka> that's cheating
15:31:39 <Gracenotes> useful.
15:31:45 <nooga_> CHEATING: D
15:31:55 <asiekierka> D for Dork?
15:31:56 <asiekierka> xD
15:33:26 <asiekierka> What about Choon?
15:33:30 <asiekierka> This would be cool
15:33:34 <asiekierka> a game that you can listen to
15:33:56 <nooga_> erm
15:34:04 <asiekierka> well
15:34:07 <asiekierka> AFAIK it doesn't have input
15:34:54 <nooga_> pitty
15:35:07 <nooga_> maybe you could code a guitar hero clone in choon
15:35:08 <nooga_> :P
15:35:16 <ehird> AnMaster: you should buy one of those usb things that look at the ambient light and colour-collaborate the monitor automatically
15:35:16 <asiekierka> but it does not have input
15:35:18 <ehird> right up your alley :P
15:36:46 <asiekierka> Choon is cool
15:37:00 <nooga_> czun
15:37:05 <asiekierka> so we have a graphical language
15:37:07 <asiekierka> a musical language
15:37:13 <asiekierka> a text language (100's of THESE)
15:37:15 <asiekierka> a poetry language
15:37:19 <asiekierka> a recipe language
15:37:20 <asiekierka> what else?
15:37:31 <AnMaster> ehird, I have considered that in fact.
15:37:42 <nooga_> once i thought about a language in which programs would be extremely complicated curves
15:37:44 <ehird> AnMaster: hahahaha
15:37:53 <asiekierka> nooga_: I planned something like that before
15:37:54 <ehird> AnMaster: lemme guess, their accompanying software isn't gpl? :-)
15:38:04 <asiekierka> where direction changes (measured in degrees) would be commands
15:38:11 <asiekierka> I lost the specification and stuff though
15:38:16 <AnMaster> ehird, actually I didn't check the details enough yet. I just saw them mentioned.
15:38:27 <nooga_> but all in all it turned out that curve is just a representation of some simple sequence
15:38:28 <AnMaster> so I put it on the mental "check details" list.
15:38:38 <ehird> AnMaster: most of them only support windows/mac
15:38:48 <ehird> since all photographers use those :P
15:38:58 <asiekierka> nooga_: It couldn't go left, too
15:39:05 <asiekierka> so you needed to use NOP commands to get back on track
15:39:06 <asiekierka> :P
15:39:18 <ehird> there should be a music language whose loops repeat in the output
15:39:19 <ehird> like [abc]
15:39:19 <ehird>
15:39:29 <ehird> [abc]abc]abc]abc]abc in the music output
15:39:36 <ehird> so your program loops have to sound musical
15:39:53 <asiekierka> hey
15:39:57 <asiekierka> i found it!!!
15:40:03 <asiekierka> It was called "Audioform"
15:40:19 <asiekierka> and I even found the specifications AND links to example programs
15:40:20 <nooga_> chloroform?
15:40:39 <asiekierka> no, Audioform
15:40:46 <AnMaster> damn parallax error
15:40:47 <AnMaster> :/
15:40:51 <AnMaster> can't make a good panorama
15:42:04 <asiekierka> lemme find the equalivment in the #esoteric logs on tunes
15:42:15 <nooga_> i go to party today's evening
15:42:39 <asiekierka> http://tunes.org/~nef/logs/esoteric/09.04.05
15:42:41 <AnMaster> fizzie, any idea why hugin decided to show one of the images upside down in the control point tab?
15:43:15 <AnMaster> fizzie, it was trying to use some birds that were flying in the auto control point thing, so I couldn't use that.
15:43:15 -!- MizardX has quit (Read error: 54 (Connection reset by peer)).
15:43:20 <asiekierka> starts on 04:33:18
15:43:30 -!- MizardX has joined.
15:43:35 <asiekierka> and it was both angle and length
15:43:44 <asiekierka> angle could be anything from -90 to 90 degrees
15:43:48 <asiekierka> as in
15:43:49 <asiekierka> in the code
15:44:01 <asiekierka> length could be anything
15:44:11 <asiekierka> ok, going away
15:44:14 -!- asiekierka has changed nick to asie[away].
15:47:59 <AnMaster> ehird, you want to see this (screenshot, safe for work, but quite high TDW factor...) http://omploader.org/vMXRxdA
15:48:04 <AnMaster> fizzie, any idea about it?
15:48:12 <ehird> AnMaster: "TDW factor"?
15:48:17 <ehird> That's the stupidest thing I ever did heard.
15:48:18 <AnMaster> The Daily Wtf
15:48:22 <ehird> I know
15:48:25 <ehird> You mean "WTF factor".
15:48:43 <ehird> AnMaster: Anyway, that image is indeed WTFy; it's bloody Raleigh theme!
15:48:46 <ehird> And over-hinted text!
15:48:57 <AnMaster> ehird, that would work too... But <retcon>with TDW is quite WTF-y itself, it ends up meaning the same</retcon>
15:49:15 <AnMaster> ehird, ignoring that... what about one image being upside down
15:49:18 <ehird> "with TDW is quite WTF-y itself" ← worst english of the year 2009 :)
15:49:29 <ehird> also, pfft, that's completely normal. i think
15:49:36 <AnMaster> ehird, s/with/due to/
15:49:42 <AnMaster> ehird, why is it normal?
15:49:46 <ehird> AnMaster: and is→being
15:49:48 <ehird> and it's not.
15:49:54 <ehird> "with TDW being quite WTF"
15:56:45 <AnMaster> ehird, s/with/what with/ would also work
15:56:55 <ehird> yse.
15:56:56 <ehird> yes.
15:57:00 -!- ais523 has joined.
15:58:31 <fizzie> AnMaster: No idea why it's upside-down; generally those things have to do with some sort of confusion in auto-rotating images based on the EXIF orientation tag.
15:59:10 <nooga_> AnMaster: you choose wrong points for comparison
15:59:43 <AnMaster> nooga_, I didn't... autopano-sift-c did
15:59:51 <nooga_> uh
16:00:04 <nooga_> saft med skinka
16:00:24 <fizzie> Is it still upside-down even if you just "load images" and don't run the automagic point-finding?
16:00:34 <AnMaster> fizzie, hm will try.
16:01:16 <AnMaster> fizzie, np
16:01:17 <AnMaster> no*
16:01:29 <fizzie> Heh, that's funny.
16:02:43 <fizzie> Well, you can run autopano-sift-c manually without that one image, open the generated .pto in Hugin, add that image-it-gets-confused-about and set up control points for it manually. (Or just skip the whole autopano-sift-c part, that's what I've been doing since I think it's rather fun to pick points.)
16:02:49 <ehird> i want a really high dpi display
16:02:55 <AnMaster> fizzie, about half of the images were upside down
16:02:57 <ehird> it'd be cool
16:03:17 <AnMaster> fizzie, not good photo weather: strong wind, raining...
16:03:31 <fizzie> AnMaster: Oh. Heh, that's rather strange. Admittedly I don't have much experience with the automatic point-select-o-tron since I haven't really been using it.
16:03:52 <AnMaster> fizzie, anyway the autopano thing is useless... It tries to match birds too. Except the birds were flying. So their positions doesn't match between the photos
16:04:06 <fizzie> Yes, well, it doesn't have an AI in it.
16:04:38 <AnMaster> fizzie, true. Would be useful to have some way to say "hey, don't look for control points here, it is birds or stuff that is moving between the pictures!"
16:04:43 <AnMaster> to mark some area like that
16:04:59 <fizzie> Anyway, you can edit & remove the points it selects manually, afterwards. (Though if it manages to turn images upside-down that's no good.)
16:05:05 <ehird> you know what's annoying?
16:05:14 <ehird> there aren't any resolutions defined above QSXGA - 2560x2048
16:05:40 <ehird> oh
16:05:42 <ehird> there is
16:05:45 <ehird> up to WHUXGA
16:05:47 <ehird> 7680x4800
16:05:57 <ehird> which is weird 16:9
16:06:18 <ehird> wait no
16:06:19 <fizzie> There are "defined" digital-film resolutions too, though those really aren't that relevant.
16:06:28 <ehird> it's uh 8:6 maybe
16:06:30 <ehird> dunno
16:06:37 <pikhq> ehird: The 16:9 of that is 7680:4320.
16:06:45 <pikhq> Which is 4xHD.
16:06:49 <ehird> "that can support a resolution up to 7680 × 4800 pixels, assuming a 16:10 aspect ratio."
16:06:53 <ehird> fuck 16:9
16:07:03 <fizzie> WHUXGA is 16:10.
16:07:07 <ehird> WQSXGA3200×204825:161.56256,553,600
16:07:09 <ehird> tha's nice
16:07:15 <ehird> fizzie:
16:07:15 <ehird> WHUXGA7680×48008:51.636,864,000
16:07:16 <ais523> 16:10 = 8:5, which is would be rather unusual
16:07:19 <ehird> wikipedia's list
16:07:23 <ehird> oh
16:07:23 <ehird> i see
16:07:28 <ehird> ais523: 16:10 isn't unusual
16:07:31 <fizzie> Yes, well, 16:10 equals 8:5.
16:07:31 <ais523> non-widescreen is 16:12, by comparison
16:07:35 <ehird> it's what most computer widescreen displays use
16:07:42 <ehird> 1920x1200 etc
16:07:42 <ais523> ah, good point
16:07:45 <ais523> this one does too, I think
16:07:51 <ehird> but saying it as 8:5?
16:07:53 <ehird> that's weird
16:08:02 <ehird> WQUXGA3840×24008:51.69,216,000
16:08:02 <ais523> why? that's lowest temrs
16:08:03 <ehird> mm
16:08:05 <ehird> wanna get me some of that
16:08:08 <ais523> people say 4:3 not 16:12
16:08:08 <ehird> ais523: because nobody says it that way
16:08:14 <ehird> 16:10 is in analogy with 16:9
16:08:24 <ais523> but nobody says 16:12!
16:08:26 <ehird> http://en.wikipedia.org/wiki/16:10
16:08:27 <ehird> wp agrees
16:08:34 <ehird> ais523: yes, but 4:3 isn't directly related to another format
16:08:45 <ehird> 16:10 is "widescreen tv with a bit more vertical space for computery stuff"
16:09:20 <pikhq> Like the movie-playing chrome THAT I DON'T HAVE!
16:09:34 <fizzie> One can put subtitles there, though.
16:09:40 <ehird> pikhq: Because monitors are just for TV.
16:09:47 <ehird> I never code. I do not have toolbars. I never overlap windows.
16:09:48 <pikhq> fizzie: True.
16:09:50 <ehird> I do not have menus or task bars.
16:10:01 <ehird> My eyes are slits that cannot move up or down.
16:10:06 <ehird> Verily, 16:9 is appropriate for me.
16:10:10 <ehird> For anyone sane, though, 16:10
16:10:11 <ehird> 's nice.
16:10:51 <pikhq> Yes, all 50 pixels make *all* the difference, and justify making an aspect ratio different from everything else.
16:10:54 <pikhq> *Right*.
16:11:08 <ehird> pikhq: Different from everything else ... except most widescreen computer displays.
16:11:12 <ehird> Yep, tiny market, that.
16:11:25 <FireFly> Meh, I never liked widescreen, I'll stick to my 4:3
16:11:27 <ehird> pikhq: Fact is, 16:9 is just too wide and not tall enough for comfortable computer usage.
16:11:38 <ehird> 16:10's a sweet spot.
16:12:20 <ais523> 8:10 is a good shape for a text editor, I think
16:12:22 * pikhq is shocked that 1/4" is a big difference between "too wide!!!" and "sweet spot".
16:12:27 <ais523> I normally set Emacs to tile two files side by side
16:12:34 <ehird> pikhq: psychologically, yes.
16:12:39 <AnMaster> ok the panorama is much much better now
16:12:42 <ehird> ais523: the golden ratio! For everything!
16:12:52 <AnMaster> still one huge mismatch
16:12:54 <AnMaster> but way fewer
16:12:57 <ehird> Give us our 1:1.6180339887 screens!
16:13:10 <ais523> that would be 16:just less than 10
16:13:14 <fizzie> I don't know where the "50 pixels" comes from; for the "normal" resolution it's the difference between 1080 lines and 1200 lines, which is 120 pixels; that's not much either, but it's not completely trivial.
16:13:39 <fizzie> I rather like 1920x1200 because it has the same height as 1600x1200 but you can still stick 1080p video on it without too much worrying.
16:13:41 <ehird> 32" tall 20" wide displays would be the golden ratio
16:13:43 <ehird> which is ridiculous
16:13:44 <AnMaster> hi ais523
16:13:48 <ehird> fizzie: 1600x1200 isn't much used.
16:13:53 <ehird> 1600x1050 is more common.
16:14:24 <fizzie> 1600x1200 is what used to be the sensible 20" 4:3 TFT size.
16:14:29 <ehird> ah
16:14:45 <AnMaster> ais523, you may want to look at http://omploader.org/vMXRxdA (safe for work and worked around the issue, but quite funny, the points were placed with the "auto" feature)
16:14:58 <ehird> ais523: did you know that the gnome UI guidelines actually tell you to keep a window size ratio of approximately 1:φ?
16:15:06 <ehird> AnMaster: you keep saying safe for work; when do we get the pornoramas?
16:15:19 <ais523> ehird: I didn't know that
16:15:29 <ehird> ais523: they're rather OCD :)
16:15:43 <fizzie> AnMaster: Actually I don't think it matters *that* much that it's upside-down. The points themselves look more or less sensible (from a quick glance) so the position-optimizer would've just turned it around when composing the actual panorama.
16:15:57 <AnMaster> ehird, ... the panorama... as soon as it looks good. Currently working on fixing a sharp edge at one place
16:16:04 <ehird> AnMaster: Pornorama.
16:16:09 <ehird> Since you keep noting every image is SFW.
16:16:12 <AnMaster> ehird, I saw and ignored
16:16:20 <ehird> I can only deduce that there must be some images you will post that are not SFW.
16:16:43 <AnMaster> ehird, heh, actually it is just that you seem to distrust that upload site
16:17:04 <AnMaster> main reasons I use it are: 1) any file format 2) no login 3) handy command line tool.
16:17:15 <ehird> AnMaster: Can you blame me, though?
16:17:15 <AnMaster> fizzie, it seems to work much better once I removed the points it placed on the birds, and added some on the mountains
16:17:16 <ehird> Q: Why?
16:17:16 <ehird> A: Omploader was originally created to become the ultimate shock site hosting service by Brenden Matthews and David Shakaryan
16:17:21 <AnMaster> currently it get mountains wrong
16:17:25 <ehird> and iirc their hotlinking image is either a shock one or "I'M GAY" or something
16:17:38 <ehird> http://omploader.org/gay.png
16:17:42 <ehird> "YOU ARE GAY, LOLS!"
16:17:55 <ehird> i mean, it doesn't exactly reek of professionalism and a haven of technical images.
16:18:47 <AnMaster> fizzie, how do you fine tune the horizon and such, you mentioned something about it.
16:18:49 <AnMaster> can't find where
16:21:56 <nooga_> hahha
16:22:09 <nooga_> today i encoutered a gay pride parade
16:22:45 <fizzie> AnMaster: There's some sort of very unintuitive thing in the "preview panorama" window. I'm not sure how it works; there's some automatic thing, and then you can click with left and right buttons at the image and it does some sort of rotations and translations.
16:22:59 <AnMaster> fizzie, uh...
16:23:09 <fizzie> On the other hand, if you then click "optimize" again it loses those twiddlings. It's a bit strange.
16:24:25 <fizzie> Anyway, you're supposed to left-click on the center of the image, and then right-click any point on the horizon, and it'll rotate it so that it is straight.
16:24:28 <AnMaster> fizzie, http://omploader.org/vMXRxdw
16:24:35 <AnMaster> fizzie, one major issue left there
16:24:45 <fizzie> But if you have actually wavy horizon (and not just tilted one) you'll probably need to add some horizontal-line control points or something.
16:24:56 <AnMaster> fizzie, and yes, I have mountains
16:24:57 <ehird> AnMaster: those clouds aren't in line
16:25:02 <ehird> also the colours don't match
16:25:03 <AnMaster> ehird, agreed
16:25:04 <ehird> first change
16:25:06 <ehird> you're welcome :P
16:25:10 <fizzie> Yes, well, I mean wavy-horizon that you don't want to straighten.
16:25:11 <ehird> AnMaster: oh and the last change
16:25:13 <ehird> is totally off.
16:25:18 <ehird> wait
16:25:19 <ehird> the second-last
16:25:26 <ehird> last one needs some colour fixing
16:25:31 <AnMaster> ehird, yes that is the main issue I mean atm
16:25:36 <AnMaster> the sharp edge
16:25:39 <ehird> :P
16:25:44 <fizzie> That's also just the preview window; the actual panorama will not look like that, I think.
16:25:52 <fizzie> Since it does quite a lot more math in blending, optimizes seams and so on.
16:26:01 <AnMaster> fizzie, how useless...
16:26:25 <AnMaster> so should I export, then check if the last seam actually looks good instead?
16:26:26 <fizzie> It takes quite some time to compute the final result, that preview is just that you can see how it arranges the source pictures there.
16:26:35 <fizzie> And I guess it should show the exposure-fixups and such there too.
16:26:56 <AnMaster> ehird, the colour difference doesn't show on this monitor
16:27:01 <AnMaster> so can't do much about it
16:27:09 <AnMaster> yes it is a sucky monitor
16:27:16 <AnMaster> well, I see one place
16:27:23 <ehird> AnMaster: well, it's really bad
16:27:24 <AnMaster> first and second pic
16:27:27 <ehird> very noticable
16:27:30 <ehird> well
16:27:34 <ehird> less so on the last one
16:27:35 <ehird> AnMaster: but
16:27:40 <ehird> AnMaster: second left is bluey
16:27:40 <AnMaster> ehird, between first and second there is a color issue yes
16:27:44 <ehird> the previous one is purpley
16:27:56 <ehird> the second-last-to-last... the last is more purpley/pinky too, than the second-to-last
16:28:00 <AnMaster> ehird, there is a place to fix it up
16:28:06 <AnMaster> but I want to get positions right first
16:28:19 <ehird> right.
16:28:24 <ehird> 16:26 fizzie: It takes quite some time to compute the final result, that preview is just that you can see how it arranges the source pictures there.
16:28:24 <ehird> 16:26 fizzie: And I guess it should show the exposure-fixups and such there too.
16:28:31 <ehird> :p
16:29:07 <fizzie> (I'm also mostly away; got the food supplies bought but now I need to prepare and consume them. Actually you're making me burn them with all this talk.)
16:29:13 <AnMaster> ehird, also I did try to tune them after each other by setting the same white balance for all. Since time delta between them were about 4-5 seconds
16:30:44 <fizzie> The more overlap you have, the better result that final stitcher-tool can do, too. I think I used about 10-20 source images mostly.
16:31:22 <fizzie> (No, really, away!)
16:31:45 <ehird> fizzie: COME BACK
16:47:07 <AnMaster> fizzie, yeah, couldn't take that many there
16:47:16 <AnMaster> too bad weather + too slow memory card
16:47:36 <fizzie> Okay, came back.
16:47:53 <AnMaster> "Precondition violation!
16:47:54 <AnMaster> BasicImage::upperLeft(): image must have non-zero size.
16:47:54 <AnMaster> "
16:47:56 <AnMaster> huh
16:48:06 <fizzie> Huh. I haven't seen anything like that.
16:48:17 <AnMaster> fizzie, it segfaulted afterwards
16:48:41 -!- Corun has quit (Read error: 110 (Connection timed out)).
16:48:54 <fizzie> Strange. Is this the stable 0.7 or the 0.8 release-candidate? :p
16:49:00 <AnMaster> fizzie, the former
16:49:08 <fizzie> Oh, even stranger.
16:49:29 <AnMaster> I did manage to open tiff after I upgraded exiv2 from 0.18.0 to 0.18.1
16:51:05 <fizzie> Speaking of hugin, I did a couple of perspective-correction tests with it too. http://zem.fi/g2/d/7368-2/p1030278.jpg -> http://zem.fi/g2/d/7733-2/mosaic-2-perspective.jpg and http://zem.fi/g2/d/7401-2/p1030297.jpg -> http://zem.fi/g2/d/7736-2/paintings-perspective.jpg
16:51:18 <fizzie> Sorry for the ugly URLs.
16:51:44 <ehird> fizzie: lol@fat people
16:52:53 <AnMaster> anyway I think the reason for the issue with the last panel is parallax issues
16:54:24 <ehird> so I was dicking around with sum OpenGL programming, and I figured I'd try to determine my polygon 'budget' as it were. On my computer with a 2.4ghz P4 and a 128mb 6600gt, I imported a 375,000 polygon model and drew it using a vertex array compiled display list at a massive 10fps. When I ran the exact same program on a new computer with a phenom x3 720 and a 1gb 4890, it ran at 600+fps. I can only assume this is due to the 128mb of the 6600gt being inade
16:54:26 <ehird> quate to entirely contain the 375,000 polygons.
16:54:29 <ehird> Even with an old p4 and the 6600gt, my old computer was able to adequately play Oblivion, which I'm sure has more polygons than that stuck in graphics memory at a given time, so what gives?
16:54:32 <ehird> — /prog/
16:55:32 -!- Corun has joined.
16:56:16 -!- Corun has quit (Remote closed the connection).
16:56:28 <fizzie> Probably. I tried to take one linear-panorama thing of the Capri coast when in a boat sliding past it, but since there was so much depth the parallaxy effects made it pretty much impossible. There's a tutorial for making linear panoramas from things like walls and such in Hugin, but it doesn't (and can't) really work when there are things in the picture at multiple distances.
16:56:32 -!- kar8nga has joined.
16:56:44 -!- Corun has joined.
17:03:37 <AnMaster> fizzie, if you hooked up a gps to the camera, couldn't you build some sort of stereo-image using such photos?
17:06:22 <fizzie> If the positional resolution is good enough, I guess. Wasn't there that 3D stockholm that was built from aerial photographs with good-enough location information?
17:07:23 <AnMaster> heh
17:07:24 <AnMaster> bbl
17:07:33 <fizzie> I've just seen still pics (I guess it needs a Java plugin in the browser) but it looked reasonably interesting.
17:07:58 <fizzie> Messy but interesting.
17:14:12 <AnMaster> fizzie, link?
17:14:54 <fizzie> I think it's the one they talk about in http://www.ogleearth.com/2008/05/see_a_3d_map_of.html
17:15:14 <fizzie> The melting-buildings style is artsy. :p
17:22:11 <nooga_> :F
17:22:37 <ehird> fizzie: the persistence of cities?
17:23:39 -!- nooga_ has quit ("Leaving...").
17:26:23 <AnMaster> whoa
17:26:34 <AnMaster> $ convert pict1219.tif -channel R -separate red.tif
17:26:35 <AnMaster> file *** glibc detected *** convert: munmap_chunk(): invalid pointer: 0x00000000023c9ef0 ***
17:26:35 <AnMaster> ======= Backtrace: =========
17:26:37 <AnMaster> [...]
17:26:48 <ehird> http://www.google.com/search?hl=en&safe=off&client=safari&rls=en-us&q=c+%2F+3.3+GHz&aq=f&oq=&aqi=
17:26:57 * AnMaster tries to think of a way to split 16 bit images by channel if not this
17:27:02 <ehird> Light travels only 9 centimeters in one clock cycle of an i7 975 XE.
17:27:57 <ehird> The highest clock x86, iirc, was an AMD Phenom II X4 @ 6.5GHz, liquid helium cooled.
17:28:04 <ehird> Light travels 4.6cm in one of its clock cycles.
17:28:11 <AnMaster> ehird, heh...
17:28:22 <AnMaster> that kind of put things in a perspective...
17:28:50 <ehird> 28 GHz = light travels 1cm in a clock cycl
17:28:51 <ehird> e
17:28:59 <ehird> pretty amazing how fast stuff is
17:29:05 <AnMaster> -fast Increase speed for slow devices
17:29:08 <AnMaster> interesting option
17:29:11 <AnMaster> I wonder how it works
17:29:20 <AnMaster> it is from man exiftool
17:30:33 <AnMaster> ah found it
17:37:38 <AnMaster> hm... 1455895040 I think we have a big/little endian messup
17:38:10 * AnMaster wonders what tool to use to convert it
17:39:05 <tetha> try i
17:39:14 <AnMaster> i?
17:39:18 <AnMaster> no such tool here
17:39:22 <AnMaster> bash: i: command not found
17:39:28 <AnMaster> anyway I used erlang to do it
17:39:34 <AnMaster> and it isn't a simple messup like that
17:53:28 <GregorR> Any number less than 2 billion (and greater than 0, for signedness) has no endian problem :P
17:56:19 <AnMaster> GregorR, it might be 16-bity
17:56:20 <AnMaster> bit*
17:56:23 <AnMaster> I'm not sure
17:56:29 <AnMaster> or 64-bit
17:56:51 <AnMaster> GregorR, I seriously have no clue. Just that exiv2 somehow messed up the EXIF values
17:57:40 <GregorR> 16-bit can't get that high, and 64-bit that is most CERTAINLY not an endianness error, the first four bytes are 0!
17:58:01 -!- MizardX has quit (Read error: 60 (Operation timed out)).
18:01:22 <Deewiant> Here's what I use to convert endianness in dobelx64's makefile: printf "0: %016x" $n | xxd -r | od -vAn -tdL
18:01:46 -!- MizardX has joined.
18:03:01 <Deewiant> The full snippet to go from integer to binary with flipped endianness: printf "0: %016x" $$sz | xxd -r | od -vAn -tdL | xargs printf "0: %016x" | xxd -r
18:05:44 -!- MizardX has quit (Read error: 54 (Connection reset by peer)).
18:06:04 -!- MizardX has joined.
18:08:53 <GuestShadowSkunk> http://www.youtube.com/watch?v=eglQ-S0MPkE
18:09:02 -!- GuestShadowSkunk has changed nick to Slereah.
18:19:19 <AnMaster> Deewiant, you mean: <<Result:32/integer-little>> = <<Old:32/integer-big>> ? ;P
18:20:16 <Deewiant> AnMaster: No, I mean something that can be done without non-POSIX dependencies :-P
18:23:34 <AnMaster> heh
18:23:50 <AnMaster> Deewiant, is this "xxd" really POSIX?
18:24:11 * ais523 wonders about sorting algorithms in Rube
18:24:24 <AnMaster> no
18:24:30 <AnMaster> Deewiant, xxd isn't POSIX
18:24:33 <AnMaster> so fail
18:25:40 <ehird> od is posix.
18:29:31 <Deewiant> It isn't? Darn.
18:29:35 <AnMaster> ehird, yes, but not xxd
18:29:37 <Deewiant> Then, challenge: make it POSIX.
18:29:46 <ehird> Deewiant: od
18:29:57 <Deewiant> ehird: I already use od.
18:30:02 <Deewiant> Tell me how to use it better.
18:30:03 <ehird> Deewiant: od can replace xxd.
18:30:15 <AnMaster> well I would try to, if I knew what xxd did
18:30:19 <AnMaster> I don't have xxd here
18:30:20 <AnMaster> so no idea
18:30:22 <ehird> hexdump.
18:30:30 <Deewiant> AnMaster: There, it reads a hexdump and translates it to the binary.
18:30:34 <ehird> AnMaster: your system isn't posix
18:30:37 <ehird> as it doesn't have vi(1).
18:30:41 <AnMaster> ehird, it does
18:30:46 <ehird> you said it didn't
18:30:50 <AnMaster> I have it symlinked to busybox vi
18:30:53 <AnMaster> ehird, it didn't
18:30:59 <AnMaster> it does since a few months
18:31:08 <ehird> why
18:31:26 <ais523> possibly for people sshing in
18:31:27 <AnMaster> ehird, since you complained so much.
18:31:38 <Deewiant> Could've sworn that xxd was part of coreutils or something, meh.
18:31:39 <ais523> I generally use vi at the other end of an ssh link
18:31:55 <AnMaster> ais523, I tend to use nano or emacs then
18:31:59 <AnMaster> depending on what I edit
18:32:00 <ehird> AnMaster: Because you're obsessive-compulsive about everything being posix and no non-posix systems being allowed.
18:32:08 <ais523> AnMaster: can you rely on emacs being at the other end of the link?
18:32:13 <AnMaster> ehird, no. Just to make you shut up
18:32:18 <ais523> nowadays you can normally rely on nano, except on embedded systems
18:32:19 <AnMaster> seems it didn't work though
18:32:29 <ehird> AnMaster: I complained because you're obsessive-compulsive about everything being posix and no non-posix systems being allowed.
18:32:35 <ehird> Reading comprehension failure.
18:33:02 <AnMaster> ehird, what you said could be interpreted more than one way.
18:33:07 <AnMaster> bbl
18:33:14 <ehird> Yes. One of those ways is retarded and nonsensical, the other is not.
18:33:18 <ehird> Next time on human communication!
18:33:26 <Deewiant> You're missing the point here, guys: do xxd -r with od
18:33:43 <ehird> Deewiant: just use sed.
18:33:47 <ehird> it's tc and ioc
18:33:51 <ehird> (io complete)
18:35:25 <asie[away]> back
18:35:37 -!- asie[away] has changed nick to asiekierka.
18:36:01 <AnMaster> ehird, Deewiant: awk is too
18:36:06 <AnMaster> and it is POSIX as well
18:36:29 <ehird> sed is nicer
18:36:42 <AnMaster> ehird, depends on what you are doing
18:36:49 <AnMaster> I suspect awk may be nicer in this case
18:36:52 <AnMaster> but I may be wrong
18:40:44 -!- zzo38 has joined.
18:42:10 <lifthrasiir> Colloquy sucks. I just switched to LimeChat 1.0, and it feels great.
18:42:46 <Deewiant> ehird: How do I sed [0-9]+ to the equivalent binary
18:43:30 <ehird> Deewiant: something like (perl) s/[0-9]{8}/eval "\"\\$0"\"/e
18:43:43 <Deewiant> perl isn't sed.
18:44:23 <ehird> Deewiant: 'Twas example.
18:44:32 <ehird> Deewiant: Anyway, a C compiler is posix.
18:44:34 <ehird> Use one of them.
18:45:15 <zzo38> Is there some reason you want it sed
18:45:25 <ehird> posix
18:45:34 <Deewiant> I don't want to depend on C :-/
18:46:01 <bsmntbombdood> i saw someone write "tr 'x' 'y'|tr 'z' 'a'|tr 'f' 'b'...."
18:46:01 <GregorR> Everyone depends on C.
18:46:07 <bsmntbombdood> i was like wtf
18:46:35 <zzo38> O, I thought you just wanted to do it with sed just so you can figure out how (not to make a real program)
18:47:10 <Deewiant> GregorR: Only incidentally.
18:49:31 <zzo38> I invented esolangs for calling itself recursively, time travel, queues to references to queues, dividing by zero, programs running by continuous equations, code golfing, and many others.
18:50:32 <ehird> Yes.
18:50:34 <ehird> Yes you did.
18:50:36 <ehird> Now how's that relevant?
18:51:03 <zzo38> I also want to know what kind of thing you invented too.
18:51:17 <zzo38> Is Q-Ref turing complete? I think it probably is because it can emulate cyclig tag
18:52:28 <zzo38> And, does Burro really form a group if the reverse of conditions cannot be used by itself?
18:52:46 <zzo38> I know Revaver2pi is definitely a group if you don't use commands with exclamation mark
18:53:14 <fizzie> bc is POSIX (though I'm not sure if the "ibase=" thing is defined there), so you can use a prettified form of: echo -e "ibase=16\n"`printf "%016X" 12345 | rev | sed -e 's/\(.\)\(.\)/\2\1/'` | bc to do endian-reversal. ... except that actually rev doesn't seem to be POSIX. Oh well.
18:53:55 <zzo38> O, so you want it POSIX compliant
18:54:37 <Deewiant> fizzie: So how do I go 8-byte integer -> 8 bytes
18:54:55 <fizzie> Oh, you wanted that too.
18:55:04 <Deewiant> It needs to be binary so I can give it to dd.
18:55:23 <zzo38> dd/sh is good enough to write programs in
18:55:27 <fizzie> I think I had something, but can't remember. Our sauna-time starts in five minutes, so no luck there.
18:56:13 <Deewiant> fizzie: tac is POSIX, isn't it?
18:56:57 <Deewiant> sed | tac = rev
18:57:02 <Deewiant> Correction
18:57:05 <Deewiant> sed | tac | sed = rev
18:57:28 <zzo38> I invented meta-complex numbers. A rank 2 meta-complex number can be represented by tensor multiplying matrix representation of two complex numbers.
18:58:05 <zzo38> A rank2 meta-complex has 4 parts, realreal, realimaginary, imaginaryreal, and imaginaryimaginary.
18:58:16 <zzo38> Those can be the numbers at the top row of the matrix
18:58:51 <zzo38> And meta-complex numbers are commutative.
18:59:00 <Deewiant> So what /is/ a meta-complex number
18:59:10 <Deewiant> Other than a 2x2 matrix
18:59:15 <zzo38> I just explained what it is, isn't it?
18:59:18 <Deewiant> No, you didn't
18:59:23 <Deewiant> You said what it can be represented as
18:59:52 <lifthrasiir> zzo38: is that equivalent to C^4, for rank 2?
19:00:00 <lifthrasiir> where C is a set of complex number, of course.
19:00:05 <zzo38> Just like complex numbers can be represented by a 2x2 matrix of reals a+bi --> [a,b;-b,a] then meta-complex numbers can be represented by matrix of real numbers, according to the rank of meta-complex numbers.
19:00:50 <lifthrasiir> is... that commutative?
19:01:12 <zzo38> Meta-complex numbers of any rank are commutative
19:01:29 <zzo38> Of rank 0 it is just like real numbers, of rank 1 it is just like complex numbers.
19:01:39 <zzo38> And I have proven it is commutative
19:02:16 <lifthrasiir> zzo38: aren't they represented as matrix? so you mean its multiplication IS NOT EQUAL TO matrix multiplication?
19:03:00 <zzo38> And, I have figured out, that as well as transposing a matrix represent a complex number to conjugate it, doing [0,1;1,0]x[0,1;1,0] where x is the representation of complex number, also works
19:03:32 <zzo38> No, their multiplication is matrix multiplication when they are represented as a matrix. And, because of the way the numbers are arranged int he matrix, it happens to be commutative.
19:03:43 <zzo38> Try it if you want to.
19:05:22 <zzo38> If a+bi and c+di are two complex numbers and you are tensor multiplying to make a meta-complex number of rank 2, then ac is the realreal part, ad is the realimaginary part, bc is the imaginaryreal part, and bd is the imaginaryimaginary part.
19:05:56 <zzo38> The result matrix is [ac,ad,bc,bd;-ad,ac,-bd,bc;-bc,-bd,ac,ad;bd,-bc,-ad,ac]
19:05:59 -!- FireFly has quit (hubbard.freenode.net irc.freenode.net).
19:05:59 -!- Gracenotes has quit (hubbard.freenode.net irc.freenode.net).
19:06:26 <zzo38> A rank 3 meta-complex number can then be represented as a 8x8 matrix
19:06:37 -!- FireFly has joined.
19:06:37 -!- Gracenotes has joined.
19:07:56 <zzo38> So, that's how meta-complex numbers work.
19:10:26 <zzo38> You can figure out square-roots and stuff just like complex numbers. If the realimaginary part is 1 and the rest are 0 then you can still square it to get negative one (you can try it)
19:11:27 <zzo38> You also get -1 if the imaginaryreal part is 1 instead
19:12:20 <zzo38> If the imaginaryimaginary part is 1 and you square it, you get positive one.
19:12:38 <ais523> zzo38: how many nth roots does 1 have?
19:13:30 <zzo38> I'm unsure because I haven't checked.
19:14:00 <zzo38> But that is a good question anyways.
19:14:52 <GregorR> I want to find an excuse to write a library called liboveslave. So you link to it by typing -loveslave
19:15:43 <ehird> GregorR: liburk
19:15:48 <ehird> liboser
19:15:53 <ehird> libuck
19:15:56 <zzo38> Obviously meta-complex numbers do not have the application of quaternions and such things as that, but you might use it for different kind of ideas
19:15:58 <ehird> libittle
19:16:04 <FireFly> libib
19:16:27 <Deewiant> libinux
19:16:29 <GregorR> libube
19:16:39 <Deewiant> libol
19:17:30 <GregorR> libollipop
19:17:33 <ehird> libook
19:17:43 <Deewiant> libunge
19:17:53 <Deewiant> libung
19:18:35 <ehird> libduck
19:18:36 <ehird> Lduck!
19:18:36 <FireFly> Are we done with s/^l/lib/'ing?
19:18:43 <GregorR> FireFly: No
19:19:03 <ehird> libbutt
19:19:23 <lifthrasiir> libibrary.
19:19:34 <Deewiant> libifthrasiir
19:19:40 <lifthrasiir> woah.
19:19:44 <GregorR> libament
19:19:49 <Deewiant> libigament
19:19:51 <ehird> libeonidas
19:19:54 <lifthrasiir> well, is there libiberty already? :p
19:19:59 <ehird> yes
19:20:02 <ehird> gnu guys made it :P
19:20:08 <ehird> libobotomy
19:20:09 <Deewiant> liback
19:20:16 <GregorR> YES
19:20:18 <GregorR> libobotomy
19:20:20 <GregorR> I'm making it
19:20:22 <lifthrasiir> lib_
19:20:33 <Deewiant> lib\
19:20:39 <ehird> GregorR: :D
19:20:52 <ehird> GregorR: libottleinfrontofme
19:21:10 <Deewiant> lottle?
19:21:18 <Deewiant> libittle
19:21:53 <ehird> Deewiant: "I'd rather have a bottle in front of me than a frontal lobotomy."
19:22:21 <Deewiant> Right
19:22:25 <GregorR> libink_no_libraries
19:22:33 <Deewiant> :-D
19:22:56 <lifthrasiir> libol.
19:22:58 <lifthrasiir> -lol
19:22:59 <Deewiant> libosing_my_mind
19:23:03 <ehird> lifthrasiir: GregorR said that before
19:23:14 <Deewiant> I said it last
19:23:17 <GregorR> No, somebody else did.
19:23:19 <GregorR> Yeah, Deewiant
19:23:22 <lifthrasiir> aargh, that was Deewiant... anyway.
19:23:22 <ehird> Kay ;P
19:23:23 <ehird> *:P
19:23:40 <GregorR> I like the idea of a library that looks like a command to gcc, like libink_no_libraries
19:23:55 <Deewiant> libess_optimization
19:24:06 <GregorR> Holy crap, I was just typing that X-D
19:24:08 <Deewiant> Or even s/ess/east/
19:24:24 <FireFly> lib0
19:24:31 <Deewiant> And I guess s/_/-/g to make it look more like a typical option
19:25:11 <GregorR> Or libicense-gpl, libicense-mit, libicense-bsd, ...
19:25:20 <GregorR> Then you could use ldd to tell you if there are licensing problems :P
19:25:21 <Deewiant> libicense-proprietary
19:25:25 <Deewiant> :-D
19:25:30 <lifthrasiir> i once thought of lib-, for -l-, and suddenly the idea of palindrome shell program popped from my mind...
19:25:52 <ais523> without cheating and using comments?
19:25:55 <ehird> lifthrasiir: echo 'easy'#'ysae' ohce
19:25:55 <Deewiant> libdd
19:25:58 <ehird> hi ais523 :P
19:26:01 <ais523> hi ehird
19:26:03 <ehird> liblllllllllllllllllllllllllllllllllllllllllll
19:26:04 <Deewiant> libd
19:26:04 <lifthrasiir> libicense-free.
19:26:07 <Deewiant> ld -ld
19:26:08 <ehird> -llllllllllllllllllllllllllllllllllllllllllll
19:26:15 <ais523> 'lib '
19:26:21 <GregorR> ais523: lawl
19:26:22 <lifthrasiir> ehird: huh, how many l's?
19:26:26 <Deewiant> ais523: That's what I was going for with lib\
19:26:30 <ehird> lifthrasiir: A load of them.
19:26:33 <GregorR> libimit-utility
19:26:51 <ehird> "libook, I'm sorry about your program, okay?"
19:26:53 <lifthrasiir> ehird: you should symbolic-link libl.so to libll.so, liblll.so, libllll.so and so on. :p
19:26:59 <ehird> (Said by Pratchett's Librarian, clearly.)
19:27:21 <GregorR> Nonono, you have a bunch of libraries libl, libll, liblll, each of which is linked to the previous one.
19:27:26 <GregorR> So -lllll implies -llll, -lll, -ll
19:27:50 <lifthrasiir> GregorR: symbolic link fail! (by the way, is there any limit on link recursion?)
19:27:55 <Deewiant> libame-warnings
19:27:57 <lifthrasiir> ais523: and yes, i meant that
19:28:09 <GregorR> lifthrasiir: No, not SYMBOLIC link.
19:28:11 <GregorR> lifthrasiir: LINK
19:28:13 <lifthrasiir> ah
19:28:19 <GregorR> lifthrasiir: Like gcc -ll foo.c -o libll
19:28:22 <GregorR> .so
19:28:35 <lifthrasiir> ah good.
19:28:49 <ehird> "lib nanoreplicator.c"
19:29:03 <lifthrasiir> "lib -pedantic"
19:29:25 <GregorR> lifthrasiir: lol
19:29:40 <ais523> -libol
19:29:42 <ais523> *libol
19:29:54 <Deewiant> "lib -Wnone"
19:30:05 <ehird> So I just did some type system hackery.
19:30:06 <GregorR> liboch-lomond
19:30:08 <ehird> All in a day's work of Haskell
19:30:27 <Deewiant> liboch-ness
19:30:31 <ais523> lib<insert literal NUL here>
19:30:39 <ais523> although I don't know any filesystems which allow NUL in filenames
19:31:19 <GregorR> libiquid-cooled-optimizations
19:31:29 <Deewiant> libava-lamp
19:32:28 <ehird> ais523: they should all allow them
19:32:29 <ehird> stupid C
19:32:35 <ehird> also, I want to use / in a filename often
19:32:40 <ehird> a lot of songs are named with a / in them, for instanc
19:32:40 <ehird> e
19:32:44 <ais523> not even bffs allows spaces in filenames
19:32:48 <ais523> although it does allow slashes
19:32:50 <ehird> traditional unix filesystems suck.
19:32:52 <ais523> *allows NUL in filenames
19:32:59 <ehird> heck, I could deal with just nul
19:33:02 <ehird> use nul as a separator
19:33:09 <ais523> haha
19:33:09 <ehird> ofc command line tools will have a syntax
19:33:12 <ehird> but you could escape it
19:33:14 <ais523> BFFS doesn't have directories at al
19:33:17 <ais523> *all
19:33:17 <ehird> in fact, you don't need any sort of separator
19:33:22 <ehird> it's a data structure
19:33:31 <ais523> they should go Cyclexa-style
19:33:37 <ais523> and have separators that aren't in the character set
19:33:46 <ehird> wow, the sun just came out really hot
19:33:50 <ehird> how did that happen
19:33:53 <Deewiant> libinked-lists-instead-of-arrays
19:33:55 <GregorR> libemon-curry?
19:34:02 <ais523> in Cyclexa you can enter codes like \F which don't correspond to any character at all, but can be matched anyway
19:35:23 <ehird> GregorR: Deewiant: lifthrasiir: ais523: http://pastie.org/510863.txt?key=jmluhbl366cxr7po3bna
19:35:27 <ehird> thread over
19:35:49 <Deewiant> grep '^l' /usr/dict/words | sed s/l/lib/ ?
19:36:09 <Deewiant> Doesn't contain libemon-curry and hence is WORTHLESS
19:36:15 <lifthrasiir> ehird: ...
19:36:18 <ehird> Deewiant: grep '^[lL]' /usr/share/dict/words|sed 's/^[lL]/lib/'|e
19:36:37 <Deewiant> |e ?
19:36:51 <ehird> Deewiant: textmate editor.
19:36:57 <ehird> my alias.
19:37:38 <Deewiant> Hmph, heavy rain warning
19:37:46 <Deewiant> Over 20mm an hour
19:38:11 <lifthrasiir> but -lemon-curry actually means WITHOUT lemon curry, so we have to add some meaning to hyphen
19:38:35 <lifthrasiir> libike?
19:38:37 <ehird> lifthrasiir: without? how?
19:38:47 <Deewiant> ehird: - instead of +, I imagine
19:38:56 <lifthrasiir> ehird: like search engine. :p
19:39:13 <ehird> Ah.
19:39:22 <lifthrasiir> well but that'd be irrelevant since all other options use -... no, forget it.
19:39:23 -!- calamari has joined.
19:39:30 <ais523> incidentally, why/when was the usage of + for switches abandoned?
19:39:33 <ehird> lifthrasiir: it means lemon but not curry
19:39:33 <ehird> :P
19:39:39 <ehird> ais523: gnu.
19:39:48 <ais523> gnu invented it? or abandoned it?
19:39:55 <GregorR> Abandoned it.
19:39:55 <lifthrasiir> ehird: ha!
19:40:01 <ais523> do you know why?
19:40:14 <ehird> ais523: --
19:40:26 <Deewiant> ehird: ++
19:40:31 <ehird> :P
19:40:39 <Deewiant> Point being, that's not a good reason :-P
19:43:26 * GregorR wants to write libemon-curry? now.
19:43:36 <GregorR> Incidentally, everybody stopped putting the important '?' there.
19:43:39 <ais523> what would it do?
19:43:48 <Deewiant> The ??
19:43:52 <GregorR> Idonno, print random Monty Python quotes to stderr?
19:45:04 <ais523> as a library?
19:45:09 <GregorR> gcc hello.c -lemon-curry? -o hello; ./hello
19:45:49 <Deewiant> libemon&&curry?
19:48:58 -!- BeholdMyGlory has quit (Remote closed the connection).
19:49:21 -!- BeholdMyGlory has joined.
19:49:56 <Deewiant> libaw-breaking-algorithms
19:52:12 -!- BeholdMyGlory has quit (Remote closed the connection).
19:53:06 -!- BeholdMyGlory has joined.
19:53:36 -!- zzo38 has quit (Read error: 104 (Connection reset by peer)).
19:54:43 -!- BeholdMyGlory has quit (Remote closed the connection).
19:54:59 -!- BeholdMyGlory has joined.
19:55:49 -!- BeholdMyGlory has quit (Remote closed the connection).
19:56:05 -!- BeholdMyGlory has joined.
19:56:46 -!- BeholdMyGlory has quit (Remote closed the connection).
19:57:01 -!- BeholdMyGlory has joined.
20:00:22 <GregorR> That's some glorious disconnecting and reconnecting.
20:02:41 -!- MizardX has quit ("What are you sinking about?").
20:09:01 -!- Microsoft has joined.
20:09:05 <Microsoft> What should I do with this nick?
20:09:39 <ehird> Microsoft: Troll.
20:09:50 <Microsoft> Well obviously.
20:09:54 <Microsoft> But troll /what/?
20:10:08 <ais523> #csharp
20:10:10 <ais523> if it exists
20:10:16 <ehird> ais523: #c#.
20:10:17 <FireFly> #c#
20:10:19 <ehird> Microsoft: ##windows
20:10:19 <ais523> ah
20:10:26 <ais523> also, wow, you're /identified/
20:10:29 <ehird> ais523: redirects to ##csharp
20:11:02 -!- tetha has quit (Nick collision from services.).
20:11:09 <Microsoft> Yes, this is /my/ nick.
20:11:11 -!- tetha has joined.
20:11:21 * ais523 wonders who ghosted tetha
20:11:27 <ais523> there was nobody here with a remotely similar nick...
20:12:06 <ehird> tetha
20:12:08 <ehird> presumably
20:12:14 <Microsoft> I can ghost myself!
20:12:30 <Microsoft> :NickServ!NickServ@services. NOTICE Microsoft :You may not ghost yourself.
20:12:33 <Microsoft> Or not.
20:12:48 * ais523 wonders what Microsoft's old nick was
20:12:55 <ehird> ais523: GregorR.
20:12:58 <ehird> RawIRC
20:13:02 <asiekierka> Mick McMicrosoft
20:13:03 <ais523> ah, aha
20:13:04 <asiekierka> lol
20:13:09 -!- asiekierka has changed nick to IBM.
20:13:16 <IBM> that's registered? wtf
20:13:27 -!- IBM has changed nick to Nintendo.
20:13:29 -!- Microsoft has changed nick to Oracle.
20:13:32 <Nintendo> That one's ALSO registered
20:13:34 <ehird> /nick InternationalBusinessMachines
20:13:42 <Oracle> Hello, I'm from Sun Microsystems.
20:13:45 -!- Nintendo has quit.
20:14:05 <ais523> *** Microsoft is now known as Oracle.
20:14:06 <ais523> perfect
20:14:11 * Oracle goes to eat a larger and more influential company.
20:14:29 <ehird> /nick Apple
20:14:45 <ehird> ais523: Buy the new iMac. Starting at $4,382,381.
20:14:53 <ehird> With a solo-core 1GHz processor and 512MB of RAM.
20:14:54 <ais523> on the other hand, it comes with a five-nines SLA
20:14:59 <ehird> Comes with an integrated 19" screen.
20:15:10 <ehird> Apple: leading the way in 2009.
20:15:29 <ehird> ais523: Six nines is cooler.
20:15:32 <ehird> 31 seconds of downtime a year.
20:16:20 <ehird> ais523: Seven nines = 3.15 seconds.
20:16:21 <ehird> Almost pi.
20:16:30 <ehird> If you're really hardcore, eight nines.
20:16:33 <ehird> 0.3s/year.
20:16:44 <ehird> That's 6ms a week.
20:16:44 <ais523> we manage 9.9999999999% uptime!
20:16:51 <Deewiant> :-D
20:17:01 <ais523> I stole that joke from someone on Slashdot
20:17:11 <Deewiant> 20 nines of uptime, all after the decimal point
20:17:15 <ehird> i think six nines is totally possible though
20:17:17 <ehird> barring nuclear fallout
20:17:28 <ehird> 31s/year? sure, i can give you that, it'll just cost through the roof.
20:17:34 <ais523> incidentally, did you know that fibre-optic cables are really vulnerable to nuclear explosions?
20:17:36 <AnMaster> so
20:17:43 <Deewiant> Why not 100%
20:17:50 <AnMaster> I have a reasonably ok panorama now. Generating the tile
20:17:52 <AnMaster> file*
20:17:54 <ais523> they're affected from miles and miles and miles away
20:17:57 <ehird> Deewiant: well, I can get 100% barring mistakes
20:18:02 <ehird> but everyone makes mistakes
20:18:09 <ais523> I remember Slicehost talking about SLAs
20:18:15 <AnMaster> and two things before ehird complains: 1) first time I do this 2) there is parallax error in the input, nothing I can do about it.
20:18:25 <ais523> companies saying that they can manage 99.999% barring (list of everything that might go wrong)
20:18:34 <Deewiant> ehird: You think you can fix a year's worth of your mistakes in 31 seconds? :-P
20:18:55 <ehird> Deewiant: I didn't say that. I said that I think I can keep servers up for a year minus 31 seconds continuously.
20:19:11 <Deewiant> Yes, and you said the cause of that minus 31 is mistakes
20:19:19 <ehird> Not really
20:19:23 <ehird> I just said that 100% is impossible
20:19:26 <ehird> Because it requires infallability
20:19:32 <AnMaster> <fizzie> bc is POSIX (though I'm not sure if the "ibase=" thing is defined there), so you can use a prettified form of: echo -e "ibase=16\n"`printf "%016X" 12345 | rev | sed -e 's/\(.\)\(.\)/\2\1/'` | bc to do endian-reversal. ... except that actually rev doesn't seem to be POSIX. Oh well. <-- isn't dc POSIX though?
20:20:15 <ehird> Anyway, it'd involve having a data center in every continent.
20:20:24 <Deewiant> ehird: And the year minus 31 seconds also does, unless those 31 seconds include mistakes.
20:20:27 <ais523> even antarctica?
20:20:32 <ehird> ais523: Well, no.
20:21:03 <ehird> half running openbsd, half running some other stable secure OS
20:21:22 <ehird> each containing a complete mirror, on a gigantic RAID-1 per server, of all the data
20:21:37 <ehird> and with two separate sets of hardware configurations
20:21:50 <ehird> so that if there's an issue with hw or os or both, we're covered
20:22:14 <ehird> and 50 dedicated load balancers per data center
20:22:16 <ehird> or more
20:22:28 <ehird> all plugged in to massively-redundant, UPSed-up-the-wazoo power supplies
20:22:43 <ehird> with two internet connections from different companies
20:22:48 <ehird> (per data center)
20:23:04 <ehird> and all systems configured for a security by denying everything and manually allowing individual tiny bits
20:23:17 <ehird> all software in operation written in a provably-correct language, and proven correct
20:23:24 <ehird> with an ultra-stable compiler — say, compiling to ada
20:23:36 <ehird> ais523: Deewiant: ...with all of this, I'm pretty sure you could manage 31s/year.
20:24:15 <Deewiant> ehird: I maintain that 31s isn't enough to notice and correct mistakes
20:24:30 -!- Petalon has changed nick to Kolonai.
20:24:48 <AnMaster> ok
20:24:50 <AnMaster> yay!
20:24:56 <ehird> Deewiant: I can't even think of a situation where there'd be an error. We're talking space-shuttle-controller levels of attention to detail here.
20:24:58 <AnMaster> it fused it correctly in final version
20:25:11 <ehird> The most you could do with such a restricted-permissions system is maybe kick out a few servers.
20:25:15 <ehird> That would do nothing.
20:25:18 <Deewiant> ehird: So then, why not 100%. :-P
20:25:23 <ehird> Deewiant: Ass-covering.
20:25:30 <ais523> ehird: most likely error in that situation is one of the servers failing, and you turning off the ones that work rather than the one that doesn't, to fix it
20:25:31 <Deewiant> Either you have to convince me that it's 0s or that it's more than 31s
20:25:39 <Deewiant> Anything in between seems just weird
20:25:55 <ais523> presumably you can stock up the SLA over the course of years
20:25:57 <ehird> ais523: How? Also, multiple data centers, remember?
20:26:04 <ais523> you can do a 5-minute fix once every decade
20:26:09 <ehird> You could take out one data center through colossal incompetence, I guess.
20:26:14 <ehird> But all of them?
20:26:19 <ais523> 5 minutes is long enough to reboot
20:26:38 <AnMaster> ais523, I assume you would dislike an 81 MB large image?
20:26:54 <AnMaster> it is a deflate-compressed 16-bits-per-channel TIFf
20:26:56 <AnMaster> TIFF*
20:27:08 <Deewiant> Consider PNGizing it or something
20:27:09 <AnMaster> panorama from north Sweden
20:27:12 <ehird> AnMaster: Gimme.
20:27:12 <ais523> I don't get why you'd send me one of those in the first place, really
20:27:19 <ais523> and deflate? on an image?
20:27:21 <ais523> why?
20:27:28 <ehird> ais523: Because otherwise it's gigantic?
20:27:29 <AnMaster> ais523, um... you know png uses deflate too?
20:27:32 <ehird> TIFF supports deflate.
20:27:35 <AnMaster> yes
20:27:51 <Deewiant> TIFF is always huge, though :-P
20:28:12 -!- MizardX has joined.
20:28:13 <AnMaster> GIMP:
20:28:16 <AnMaster> "Warning:
20:28:16 <AnMaster> The image you are loading has 16 bits per channel. GIMP can only handle 8 bit, so it will be converted for you. Information will be lost because of this conversion."
20:28:30 <AnMaster> but they are working on it
20:28:38 <ais523> still, 48-bit colour is surely far too much
20:28:53 <ais523> 64-bit if you added alpha!
20:29:00 <AnMaster> ais523, it does have alpha
20:29:10 <AnMaster> ais523, for where there is no panorama data
20:29:14 <ais523> ag
20:29:15 <ais523> *ah
20:29:21 <ais523> I was wondering how a photo could have alpha...
20:29:27 <ehird> ais523: so how do you take out multiple of those datacenters at once?
20:29:28 <AnMaster> ais523, and yes it is too much. The camera produces 12 bits per channel
20:29:30 <AnMaster> in the raw image
20:30:01 <AnMaster> ais523, but that is, like, highly annoying to work with outside the raw format for the specific camera!
20:30:19 <ehird> $(declareMessage Client "HAVER" ["useragent", "extensions"])
20:30:19 <ehird> $(declareMessage Server "HAVER" ["hostname", "serveragent", "extensions"])
20:30:21 <ehird> lah de dah
20:30:29 <ehird> haskoli
20:32:26 <AnMaster> ais523, is 5.5 MB acceptable?
20:32:45 <AnMaster> or maybe 4.6
20:32:55 <ais523> AnMaster: I don't get why you're asking if an image size is acceptable, when I didn't ask for the image in the first place
20:33:07 -!- kar8nga has quit (Remote closed the connection).
20:33:23 <AnMaster> ais523, I think you would like it. No human activity visible for miles around.
20:33:49 <AnMaster> well, actually there was, just out of the picture (my backpack)
20:34:06 <AnMaster> oh and the canoe too, not visible in the pic either
20:35:14 <ehird> ais523: I don't know what's come over him either.
20:35:22 <ehird> Let's just pretend to download his photos.
20:35:36 <AnMaster> ehird, hey, if fizzie can show his panoramas I don't see why I can't!
20:35:49 <Deewiant> wget --delete-after
20:36:09 <AnMaster> plus I resized it now and such, so much smaller
20:36:35 <AnMaster> http://omploader.org/vMXRzYQ (4.6 MB)
20:38:43 -!- sebbu2 has joined.
20:39:03 <bsmntbombdood> AnMaster: high dynamic range is good
20:39:36 <AnMaster> bsmntbombdood, well, I don't think it is HDR as such. But not too bad for the first panorama I made.
20:39:51 <ais523> wow, Wikipedia has a list of games for which up up down down left right left right b a is a cheat code
20:40:08 <bsmntbombdood> AnMaster: more bits = more dynamic range
20:40:14 <AnMaster> and that one is downsampled to 8 bits per channel, if you actually want the original tiff, it is available on request
20:40:21 <fizzie> I resized mine to something like 4000x1000 and less-than-a-megabyte .jpgs, though. :p
20:40:47 <AnMaster> fizzie, hey, I put it at quality = 96 in gimp!
20:41:02 <ais523> not just games; apparently it also affects facebook and digg
20:41:24 <AnMaster> fizzie, but what do you think about the picture in question
20:41:25 * ehird attempts to do template haskell trickery
20:41:32 -!- sebbu has quit (Read error: 60 (Operation timed out)).
20:41:45 <pikhq> AnMaster: Make it PNG.
20:41:49 <pikhq> ;)
20:41:54 <AnMaster> pikhq, hm? it would be larger in fact
20:42:10 <AnMaster> what I could do is cut some padding at the edges it seems
20:42:13 <pikhq> Well, yes... PNG *is* a lossless format...
20:42:25 <AnMaster> pikhq, the goal here was small file
20:42:57 <fizzie> I cropped mine so that they were rectangular full-of-image-data, even though I had to discard quite a bit of actual content there.
20:43:37 <AnMaster> fizzie, that might be better... hm. But it would cut too much here
20:44:33 <bsmntbombdood> what would happen if you compressed video by take 1, 1^2, 1^2^3, 1^2^3^4, etc, and then DEFLATEd that?
20:44:46 <Slereah> BOOM
20:44:54 <bsmntbombdood> where ^ is xor and 1,2,3 are frames
20:45:00 <ehird> what's 4 then
20:45:07 <Slereah> A nigger
20:45:15 <bsmntbombdood> ...the 4th frame
20:45:29 <ehird> :)
20:45:50 <AnMaster> heh
20:46:20 <AnMaster> http://omploader.org/vMXRzZA (grumble... 5000xwhatever, 3.2 MB...)
20:46:27 <AnMaster> fizzie, prefer that
20:46:29 <AnMaster> ?
20:46:31 <ais523> bsmntbombdood: that's close to how real video compression algorithms work
20:46:38 <pikhq> bsmntbombdood: I strongly suspect that would be huge.
20:46:52 <ais523> the main difference is that they divide the image into squares, and xor each square with the other square it's most similar to
20:46:55 <pikhq> ais523: That seems to have very little with the discrete cosine transform.
20:47:01 <ais523> then you DCT the result
20:47:03 <ais523> and then compress that
20:47:18 <pikhq> Never mind. :P
20:47:44 <ais523> I think the xor happens before the DCT, but I'm not sure
20:47:53 <bsmntbombdood> ais523: i'm thinking lossless
20:48:05 <ais523> DCT /is/ lossless, in theory
20:48:28 <ehird> lossless video is love
20:48:31 <ais523> although it's common to throw away the less informative bits of the result, in which case it's lossy
20:48:32 <ehird> i want lossless 1080p video
20:48:33 <fizzie> AnMaster: It's more sensibly cut, yes. Your choice of content is harder to panoramize, there's more stuff in front.
20:48:41 <ehird> ais523: store the result as equations
20:48:45 <ehird> infinite precision
20:48:54 <ehird> i wonder how big lossless 1080p would be
20:49:00 <ehird> well one frame 1080p lossless is...
20:49:05 <ehird> 5MB? 10MB?
20:49:13 <bsmntbombdood> ehird: uncompressed 1080p is 177 megabytes/second
20:49:28 <bsmntbombdood> assuming fps of 30
20:49:31 <pikhq> Compressed 1080p is significantly less.
20:49:40 <ehird> bsmntbombdood: so 1,244GB for one movie
20:49:40 <ais523> http://kevan.org/rubicon/game.php?level=kycyfog
20:49:45 <bsmntbombdood> how good is lossless compression these days?
20:49:47 <fizzie> The stuff they use digital movie theatres is encoded each frame as a separate jpeg2000 image.
20:49:47 <ehird> i bet you could compress that to 200GB
20:49:58 <pikhq> Rather good.
20:49:58 <ehird> losslessly
20:50:06 <ehird> i just hate lossiness
20:50:10 <ehird> i'm such a packrat
20:50:29 <ehird> i'd buy 1PB of storage for like $100/mo
20:50:31 <ehird> :P
20:50:33 <ehird> i'm also a cheapskate.
20:50:45 <bsmntbombdood> http://www.google.com/search?q=1080*1920++*+24+bits+*+30+hertz+*+1+hour
20:50:48 <bsmntbombdood> google is awesome
20:50:51 <fizzie> And uncompressed 1080p is 177 MB/s only if you use 24-bit color depths; I don't think people use that in video, it's always some sort of YUV420 or whatever.
20:51:09 <fizzie> With a lower-resolution chroma channels and so on,
20:51:11 <bsmntbombdood> fizzie: lame
20:51:19 <ehird> bsmntbombdood: 1.2TB for 120 mins
20:51:23 <ehird> = 1,244GB like I said
20:51:36 <ehird> mathematica can't calculate DCTs as equations instead of values
20:51:37 <ehird> that's lam
20:51:37 <ehird> e
20:51:38 <pikhq> fizzie: Raw 1080p tends to have same-resolution chroma channels.
20:51:39 <ehird> that's ultra lame
20:51:40 <AnMaster> fizzie, that is true
20:51:48 <ehird> FREAKING LAME!
20:51:53 <AnMaster> fizzie, and it was my first time. Plus it was raining a lot
20:52:03 <AnMaster> fizzie, and strong wind
20:52:16 <AnMaster> so a lot of parallax issues to work around
20:52:18 <ehird> AnMaster: have you noticed that nobody cares that
20:52:21 <ehird> *yet
20:52:36 <AnMaster> ehird, I just replied to his comment.
20:52:44 <AnMaster> ehird, and I noticed that *you* don't care.
20:52:45 <ehird> ...what comment?
20:52:47 <AnMaster> which is very different
20:52:58 <AnMaster> ehird, you can't read scrollback either: <fizzie> AnMaster: It's more sensibly cut, yes. Your choice of content is harder to panoramize, there's more stuff in front.
20:53:01 <pikhq> fizzie: Also, it's 24-bit Y'CbCr.
20:53:02 <ehird> erm, ais523 also said he didn't ask for it, and nobody else has commented apart from fizzie
20:53:10 * AnMaster was afk
20:53:14 <pikhq> (misnomer -- YUV)
20:53:14 <fizzie> pikhq: Yes, I guess when it's raw it could be done.
20:53:19 <ehird> AnMaster: that was ages ago
20:53:32 <fizzie> (Now I am again not here.)
20:53:45 <AnMaster> ehird, as I said. I was afk.
20:54:03 <ehird> bsmntbombdood: FourierDCT[N[{1, 2, 3, 4}, 100000]]
20:54:10 <ehird> bsmntbombdood: 100,000 digit precision good enough for you
20:54:11 <ehird> ?
20:54:14 <ehird> almost lossless :P
20:54:23 <AnMaster> 21:53:09 * AnMaster was afk 21:53:19 <ehird> AnMaster: that was ages ago <-- I guess it was lag?
20:54:37 <ehird> AnMaster: no, it's just that accusing me of not reading all of the day's backlog is silly
20:54:45 <ehird> when i get annoyed at you for not reading it was just a few minutes up
20:54:46 <bsmntbombdood> wtf!
20:54:51 <ehird> bsmntbombdood: what
20:54:52 <bsmntbombdood> wolfram alpha just told me to gtfo
20:54:52 <AnMaster> ehird, um. You were here when he said that
20:54:56 <ehird> bsmntbombdood: lolwut
20:54:58 <AnMaster> so I fail to see your point
20:55:06 <ehird> AnMaster: i read every message and remember it forever!
20:55:14 <AnMaster> ehird, and it was also just a few minutes ago here
20:55:19 <AnMaster> the one that I were replying to
20:55:22 <FireFly> bsmntbombdood, literally?
20:55:39 <AnMaster> <ehird> 21:52 - 21:48 = ages
20:55:53 <ehird> oh, shut up.
20:56:00 <ehird> it was drowned in a sea of other messages
20:56:07 <ehird> and was way further back than anything I ever complain about
20:56:09 <AnMaster> ehird, ok, *that* is a valid excuse.
20:56:12 <AnMaster> the other ones weren't
20:56:32 <AnMaster> ehird, not at all, this was within one screen. But most of those your mention are more than that
20:59:06 <GregorR> LAAAAAAAAAAAL
21:01:00 -!- Oracle has changed nick to SCO.
21:01:16 <SCO> hai gais, all ur linux r belong to us
21:01:28 <ais523> aarch!
21:01:31 <ais523> *aargh!
21:01:40 <SCO> ais: Yes, ArchLinux too.
21:02:03 -!- SCO has changed nick to Intel.
21:02:11 -!- Intel has changed nick to AMD.
21:02:15 <AMD> Darn, owned :P
21:02:33 -!- AMD has changed nick to Intel.
21:02:59 <FireFly> You, or the nick?
21:03:07 <Intel> The nick.
21:03:09 <FireFly> Ah
21:03:19 <ais523> SCO was owned by Novell's lawyers
21:04:48 <Intel> Novell is owned :(
21:06:36 -!- Intel has changed nick to poop.
21:06:49 <GregorR> * Intel is now known as poop // for the logs :P
21:07:45 -!- poop has changed nick to Microsoft.
21:07:52 <GregorR> * poop is now known as Microsoft // also for the logs
21:11:11 <AnMaster> GregorR, :)
21:11:18 <AnMaster> GregorR, Long lives AMD
21:11:22 <AnMaster> err
21:11:24 <AnMaster> live*
21:11:27 * GregorR is an Intel fan :P
21:11:45 <GregorR> Which is why I'm so damned loud.
21:11:48 <GregorR> GET IT? IT'S A PUN
21:11:53 <AnMaster> heh
21:12:24 <GregorR> Anybody in here also in #opensolaris? I want to take Oracle in there, say "NOM NOM NOM", then leave, but see if anybody reacts :P
21:12:35 <ehird> joined
21:12:45 <GregorR> OK, I'll wait a few minutes :P
21:12:51 -!- Microsoft has changed nick to Oracle.
21:13:05 -!- Oracle has left (?).
21:13:19 <FireFly> Did you troll something with Microsoft, btw?
21:13:29 <GregorR> No, never thought of a worthwhile trolling :P
21:13:37 <ehird> GregorR: ##windows
21:13:43 <ehird> GregorR: announce something.
21:13:51 <GregorR> ehird: Ooooh, announce something, that's good.
21:14:03 <ehird> GregorR: Like, like, the new Zune.
21:14:04 <FireFly> "Windows 7 is now officially closed."
21:14:10 <ehird> Or Windows 8, based on Linux.
21:14:16 <ehird> GregorR: Or moving to Itanium.
21:15:23 <ehird> 21:15 Oracle has left ()
21:15:23 <ehird> 21:15 sylence: another thing: why do all my terminals (gnome-terminal, xterm, urxvt) have no color support by default? how to activate color support?
21:15:26 <ehird> 21:15 cgkades: lol
21:15:28 <ehird> 21:15 kc8bws: I can see flash clips but not watch youtube trash
21:15:30 <ehird> GregorR: a stunning reaction.
21:15:47 <GregorR> Well, somebody said "lol", that's something X-P
21:15:58 <ehird> GregorR: I bet those ##windows peeps are defensive.
21:16:01 <lament> lol
21:17:12 <GregorR> ehird: I still haven't thought of the perfect thing to say though.
21:17:47 <ehird> GregorR: Set up a webpage on an IP address mimicking the microsoft site, announcing something. Paste the link with a corporatey explanation
21:17:54 <ehird> "We are sorry to announce to our customers that, etc."
21:17:58 <ehird> "- Microsoft PR Department"
21:18:04 <ehird> IP so the domain doesn't give it away
21:18:12 <GregorR> The IP gives it away :P
21:18:19 <GregorR> (Its fakeness, that is)
21:18:21 <ehird> GregorR: Put it on microsoft.com then
21:18:28 <GregorR> No prooooooooooblem :P
21:19:14 <FireFly> [22:17:37] Microsoft is Microsoft!n=Microsof@65.183.185.209
21:19:25 <FireFly> Hm
21:20:33 <ehird> GregorR: announce an official partnership with Samba
21:20:37 <ehird> then get more and more absurd
21:21:16 <GregorR> I think something along the lines of the Microsoft-Novell thing could be fun.
21:21:37 <ehird> GregorR: Announce that MS & Novell are acquiring Samba.
21:23:54 <GregorR> <Microsoft> Microsoft is now announcing the acquisition of Novell and all of its subsidiaries. This acquisition will greatly enhance several of Microsoft's enterprise-class server products, as SUSE technologies are integrated into our new platform, Windows Linux Enterprise.
21:24:37 <ehird> :D
21:24:41 <ehird> GregorR: But wait
21:24:48 <ehird> GregorR: Add some backstory about communicating with the community
21:24:57 <GregorR> Hmmmmmmmmmmm ......
21:24:58 <ehird> eg "I was tasked to correspond with Windows communities across the web"
21:25:03 <ehird> So they don't think you're too special
21:25:06 <ehird> [note: this won't work]
21:25:31 <GregorR> That's why I think that just announcing it with no other info is fun :P
21:25:48 <ehird> GregorR: But nobody will responddd.
21:25:54 <ehird> you have to make yourself lowly :P
21:26:21 -!- kar8nga has joined.
21:27:48 <ehird> GregorR: you might get away with it. a guy's defending IE.
21:28:05 <GregorR> ehird: But your suggestion is true, so I'm writing :P
21:28:11 <ehird> 21:27 Ether_Man_: ryaxnb, yes really.. Opera has been first with virtually all the features of both IE and firefox
21:28:14 <ehird> and also Opera!
21:30:21 <GregorR> <Microsoft> As this merger will likely have a powerful influence on the Open Source ecosystem, we have chosen to ...
21:30:26 <GregorR> It just sounds so ridiculous :P
21:30:49 <ehird> GregorR: 'scuz it is. MS would never talk in an unofficial irc channel on an open source network.
21:31:00 <ehird> Especially not without any link to an official announcement, and not with the username Microsoft.
21:31:14 <GregorR> Heh :P
21:31:19 <GregorR> If I can fool just /one/ person
21:31:32 <ehird> GregorR: there's nobody that stupid on freenode
21:32:10 <ehird> 21:31 Ether_Man_: ehird, that's not really true and you should know that.. Mozilla and netscape is NOT the same project, nor is Mozilla and Firefox the same project
21:32:11 <ehird> 21:32 ehird: Ether_Man_: Mozilla is the name for the open source continuation of netscape, dude. Original mozilla's code was identical to netscape. Firefox was just a fork of Mozilla that stripped out features.
21:32:14 <ehird> let's see if idiots challenge me
21:33:25 <ehird> 21:32 Ether_Man_: ehird, not exactly no.. Do better research :)
21:33:25 <ehird> 21:33 ehird: Ether_Man_: I've followed the Mozilla project for years and have this info from authoritative sources. You are merely making unfounded assertions. Rebut me, if you wish, but this is not convincing and you are wrong.
21:33:30 <ehird> i'm psychic.
21:34:55 <GregorR> Maybe I'll just stick to the initial announcement, plus <Microsoft> If you have any questions or concerns regarding this acquisition, please contact Microsoft support at novellmerger@microsoft.com
21:35:12 <ehird> 21:34 Ether_Man_: ehird, http://en.wikipedia.org/wiki/Mozilla they reference all the info.. Just read... Please
21:35:13 <ehird> 21:35 ehird: Ether_Man_: What a copout. You are simply wrong and I don't care that you think otherwise. Or at least, if you're right, then ex-Netscape employees and the Mozilla organization don't know about it.
21:35:23 <ehird> GregorR: Troll that etherman guy please. He's an idiot.
21:35:38 <AnMaster> ehird, we are so not going to do so
21:35:38 <GregorR> Of course he is, he's on ##windows :P
21:35:50 <AnMaster> GregorR, hm that changes that a bit
21:35:51 <ehird> AnMaster: I am talking to GregorR; what are you talking about?
21:36:01 <ehird> I did not mention you at all.
21:36:16 <AnMaster> mhm
21:36:25 <AnMaster> what is your point?
21:36:55 <ehird> GregorR: Can you translate what AnMaster's talking about since 21:35 AnMaster: ehird, we are so not going to do so ?
21:36:57 <FireFly> There it goes
21:36:57 <ehird> I have no fucking clue.
21:38:02 <GregorR> No reaction WHATSOEVER :(
21:39:58 <ehird> 21:37 Ether_Man_: ehird, so basicly what you tell me is that you're a better and more authorative source than Mozilla org themselves? Righto..
21:39:58 <ehird> 21:37 ehird: Ether_Man_: Please, you are welcome to quote me sentences that disprove what I say. Could you take a few seconds to do that instead of pointing me to the same wiki page that agrees with me?
21:40:03 <ehird> 21:39 Ether_Man_: ehird, you do know what a release roadmap is I hope? And what an Initial draft means? https://wiki.mozilla.org/ReleaseRoadmap
21:40:10 <ehird> 21:39 ehird: Ether_Man_: Great, you didn't listen to me and you link to a totally irrelevant page. You're bloody Einstein, idiot.
21:40:16 <ehird> My idiotdar beeps.
21:41:14 <ais523> ehird: are you telling the truth, or trying to mislead people?
21:41:18 <GregorR> I'm sad that there was no reaction at all :P
21:41:33 <ehird> ais523: I'm telling the truth entirely; Ether_Man is a misguided idiot who cannot accept he's wrong and refuses to give arguments.
21:41:44 <AnMaster> I'm going to play with HDR tomorrow. Getting an image with both very light and very dark areas.
21:41:49 <ehird> I'm pretty damn certain exactly how the Mozilla project wen.
21:41:50 <ehird> went
21:42:17 <GregorR> ehird: I don't understand what he believes ...
21:42:25 <GregorR> ehird: That Mozilla and Firefox are (source-wise) unrelated?
21:43:06 <ehird> GregorR: I don't even know; I think he thinks that Firefox uses Gecko but wasn't a fork of Mozilla (untrue), and that Mozilla was not just initially a rebranding of Netscape (colossally untrue)
21:43:20 <GregorR> ehird: Think I should try #suse next?
21:43:25 <GregorR> (Which I guess links to #opensuse
21:43:25 <ehird> meh
21:43:26 <GregorR> )
21:43:32 <ehird> Burb.
21:43:35 <ehird> (brb)
21:45:56 <GregorR> ehird: Join #opensuse
21:47:04 <GregorR> Oh, he's brbing.
21:47:06 <GregorR> Never mind :P
21:51:34 <GregorR> lawl, they got the cavalry on #suse :P
21:52:05 <FireFly> [22:51:20] Microsoft [n=Microsof@65.183.185.209] has quit IRC:
21:52:06 <FireFly> [22:51:35] <ryaxnb> microsoft, having succesfully trolled, has left freenode
21:52:06 <FireFly> [22:51:47] virsys [n=virsys@or-67-232-64-36.dhcp.embarqhsd.net] has quit IRC: Read error: 60 (Operation timed out)
21:52:06 <FireFly> [22:51:53] <ryaxnb> he got me; i googled it
21:52:07 <FireFly> Be happy
21:52:19 <GregorR> laaaaaaaaawl
21:53:02 <GregorR> When somebody @freenode/staff gets angry, it's time to /quit :P
21:53:21 <FireFly> Heh
21:53:49 <GregorR> What do you bet my nickserv registration disappears? :P
22:01:14 -!- psygnisfive has quit (Read error: 104 (Connection reset by peer)).
22:01:29 -!- psygnisfive has joined.
22:01:41 <bsmntbombdood> i wonder when google will put a full cas into their calculator
22:02:44 <ais523> cas?
22:02:52 <Deewiant> See: wolframalpha.com
22:02:55 <Deewiant> ais523: Computer algebra system
22:02:58 <ais523> ah, ok
22:03:15 -!- psygnisfive has quit (Read error: 104 (Connection reset by peer)).
22:05:00 -!- psygnisfive has joined.
22:07:23 <GregorR> OpenID is simultaneously extremely useful and totally useles.
22:07:27 <GregorR> *useless
22:12:27 -!- psygnisfive has quit (Read error: 104 (Connection reset by peer)).
22:14:43 -!- psygnisfive has joined.
22:15:02 -!- psygnisfive has changed nick to augur_.
22:15:16 -!- augur_ has changed nick to augur.
22:21:04 -!- augur has quit (Read error: 104 (Connection reset by peer)).
22:22:28 -!- psygnisfive has joined.
22:32:40 <bsmntbombdood> bah
22:32:48 <bsmntbombdood> now i'm thinking about video compression
22:35:19 <GregorR> I enjoy Onerously Uptight Toccata more than I wish to admit :P
22:41:45 -!- Gracenotes has quit (Read error: 104 (Connection reset by peer)).
22:41:51 <ehird> GregorR: gimme a log
22:41:52 -!- Gracenotes_ has joined.
22:41:53 <ehird> i wanna see it
22:41:56 -!- Gracenotes_ has changed nick to Gracenotes.
22:42:01 <ehird> 22:07 GregorR: OpenID is simultaneously extremely useful and totally useles.
22:42:01 <ehird> 22:07 GregorR: *useless
22:42:06 <ehird> it's not useless!
22:42:08 <ehird> just underused
22:42:15 <ehird> bsmntbombdood: let's invent a lossless video compression algo
22:42:19 <GregorR> I added OpenID support to Masterpiecemachine.
22:42:33 <bsmntbombdood> ehird: i'm working on it
22:42:36 <bsmntbombdood> ehird: i've got some good ideas
22:42:47 <ehird> bsmntbombdood: tuning it for animation or real?
22:43:35 <bsmntbombdood> neither
22:43:56 <ehird> bsmntbombdood: you have to do one :P
22:44:01 <ehird> or it'll be sub-optimal
22:44:20 <ehird> bsmntbombdood: what sort of compression ratios are you looking at for movie-length (120 minutes) 1080p?
22:44:26 <ehird> from 1.2TB to hopefully...
22:44:29 <bsmntbombdood> i'm not that far yet
22:44:35 <ehird> kay :P
22:44:48 <ehird> bsmntbombdood: but what would you like? (well, in reason; no everything-becomes-1-byte stuff)
22:44:57 <bsmntbombdood> i've got no clue
22:46:25 -!- Gracenotes has quit (Read error: 104 (Connection reset by peer)).
22:50:17 <ehird> bsmntbombdood: care to share a few ideas?
22:50:19 <ehird> video compression is fun
22:50:28 <ehird> especially if you can decode it in realtime
22:50:53 <bsmntbombdood> well it goes like this
22:51:49 <bsmntbombdood> pad, break into 16*16 or whatever chunks, output an undelimited stream of them | \
22:52:35 <ehird> bsmntbombdood: I've always disliked specific chunk sizes
22:52:41 <bsmntbombdood> xor all chunks in a sliding window, sort based on hamming weight, output the lowest | \
22:52:42 <ehird> Since it leads to blocky-looking stuff lossy
22:52:42 <bsmntbombdood> gzip
22:52:50 <ehird> for lossless, not as much of a problem :LP
22:52:52 <ehird> *:P
22:52:55 <ehird> But...
22:52:59 <ehird> Dunno.
22:53:05 <ehird> for HD you'd want like 128x128
22:53:11 <ehird> bsmntbombdood: bzip2 instead of gzip?
22:53:23 <ehird> or that really fast, really good compression algorithm, forget its name
22:53:46 <ais523> lzma is slow to compress but fast to decompress
22:54:13 <bsmntbombdood> doesn't matter
22:54:30 <bsmntbombdood> let's just say LZ77
22:58:04 <ehird> bsmntbombdood: any ideas beyond that? :P
22:58:38 <bsmntbombdood> you can probably do better with some specialized entropy coding for whatever distribution it turns out you get
23:00:58 <bsmntbombdood> i'm still working out how to do the middle step
23:05:18 <bsmntbombdood> hrm
23:05:52 <bsmntbombdood> given A, A^D, D^B, B^E, E^C, you can solve for all A, B, C, D, E, right?
23:08:05 <psygnisfive> bsmntbombdood: what?
23:09:02 <bsmntbombdood> ^ is xor
23:09:10 <psygnisfive> ok
23:09:30 <psygnisfive> yes, then.
23:09:44 <psygnisfive> A^(A^D) = D, iinm
23:09:47 -!- MigoMipo has quit (Read error: 104 (Connection reset by peer)).
23:09:52 <ehird> bsmntbombdood: of course
23:10:00 <ehird> bsmntbombdood: since A^B is reversible with either A or B
23:10:06 <bsmntbombdood> yeah
23:10:08 <ehird> (just xor them :P)
23:10:12 <bsmntbombdood> ok i think i've got it
23:10:13 <ehird> xor is awesome
23:10:15 <ehird> it combines data!
23:10:16 <bsmntbombdood> let me write some code
23:10:18 -!- MigoMipo has joined.
23:10:25 <psygnisfive> xor is pretty awesome
23:10:54 <psygnisfive> it lets you compress N bits of data to N/2+1 bits
23:11:17 <psygnisfive> i think. :D
23:11:32 <psygnisfive> tho maybe not! :D
23:11:58 <bsmntbombdood> erm...
23:12:00 <ehird> psygnisfive: erm no, it doesn't save any space.
23:12:04 <ehird> ...
23:12:15 <psygnisfive> shut up stop ruining my illusions! :|
23:13:09 <bsmntbombdood> 2**64 chunks is enough for anyway right?
23:14:07 <ehird> bsmntbombdood: that's 1180591620717411303424 pixels if you use 64 pixel blocks
23:14:09 <ehird> so ehm yes
23:21:10 <ehird> bsmntbombdood: what source material are you testing this on?
23:21:19 <ehird> (you should compare it with huffyuv btw.)
23:21:22 <bsmntbombdood> ....i don't have anything yet
23:21:33 <ehird> :D
23:22:39 -!- M0ny has quit.
23:22:48 <bsmntbombdood> uuh, 26 lines
23:28:24 -!- ais523 has quit (Remote closed the connection).
23:32:14 <bsmntbombdood> ehird: i can use sse4 for the hamming weights :D
23:32:24 <ehird> bsmntbombdood: :D
23:32:25 <ehird> awesome
23:32:43 <bsmntbombdood> and see.whatever for the xoring of course
23:33:00 <ehird> bsmntbombdood: 128-bit blocks then? :P
23:33:08 <ehird> so uh, 5.3 pixels
23:33:12 <bsmntbombdood> no need
23:33:20 <ehird> bsmntbombdood: eh?
23:33:30 <bsmntbombdood> they can be as big as needed
23:33:42 <ehird> bsmntbombdood: with sse?
23:33:47 <ehird> it operates on 128 bit values
23:33:51 <bsmntbombdood> ...
23:33:57 <ehird> ...no?
23:34:39 <bsmntbombdood> it does, but you can just add
23:34:45 <ehird> well, yes :P
23:41:13 -!- kar8nga has quit (Remote closed the connection).
23:44:07 -!- MigoMipo has quit ("QuitIRCServerException: MigoMipo disconnected from IRC Server").
23:51:49 -!- inurinternet has joined.
←2009-06-12 2009-06-13 2009-06-14→ ↑2009 ↑all