00:08:05 -!- Sprocklem has quit (Quit: brb).
00:18:22 -!- Sprocklem has joined.
01:10:36 -!- Remavas has quit (Remote host closed the connection).
01:18:43 -!- doesthiswork has joined.
01:29:45 -!- GeekDude has quit (Ping timeout: 260 seconds).
01:33:21 -!- hppavilion[1] has joined.
01:36:31 -!- HackEgo has quit (Ping timeout: 240 seconds).
01:37:29 -!- HackEgo has joined.
01:52:48 -!- oerjan has joined.
02:14:32 <lambdabot> KOAK 190053Z 30013KT 10SM FEW030 FEW060 29/15 A2972 RMK AO2 SLP065 T02940150
02:14:44 <shachaf> \toodamnhigh{the temperature}
02:14:50 <lambdabot> KSJC 190053Z 30014KT 10SM FEW090 FEW120 34/15 A2970 RMK AO2 SLP055 TCU DSNT NE-E T03440150 $
02:14:55 <HackEgo> lambdabot: @@ @@ (@where weather) CYUL ENVA ESSB KOAK PAMR
02:14:57 <lambdabot> CYUL 190100Z 21013G19KT 15SM FEW040CB BKN100 BKN170 BKN240 28/20 A2956 RMK CB1AC6AC1CI1 CB TR CI TR SLP010 DENSITY ALT 2100FT \ ENVA 190050Z 24007KT 9999 FEW027 BKN057 11/09 Q1014 RMK WIND 670FT
02:14:57 <lambdabot> 26010KT \ ESSB 190050Z AUTO 25006KT 9999 NCD 18/14 Q1012 \ KOAK 190053Z 30013KT 10SM FEW030 FEW060 29/15 A2972 RMK AO2 SLP065 T02940150 \ PAMR 190053Z AUTO VRB05KT 10SM BKN070 17/07 A2999 RMK AO2
02:23:57 <shachaf> oerjan: Is that a political party?
02:24:49 * oerjan checks the bitcoin bubble
02:24:58 <shachaf> cf. https://en.wikipedia.org/wiki/Rent_Is_Too_Damn_High_Party
02:25:06 <oerjan> not quite as bubbly as last time.
02:26:00 <oerjan> there might well be a tiny party with a silly name, what do i know.
02:27:29 <oerjan> a previous election had some comedians running. although they didn't win like in iceland.
02:28:42 <oerjan> hm this year is a new election.
02:32:42 <__kerbal__> Someone should create a Birthday Party.
02:32:45 <oerjan> the norwegian cannabis party and the pirate party are the silliest names i can see
02:33:01 <oerjan> and the extinct natural law party
02:34:19 <oerjan> (which proposed basing everything on transcendental mediation)
02:49:32 -!- hppavilion[1] has quit (Ping timeout: 255 seconds).
02:57:10 <HackEgo> [wiki] [[Special:Log/newusers]] create * Copy * New user account
03:01:42 <HackEgo> [wiki] [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=52195&oldid=52190 * Copy * (+118)
03:01:49 <HackEgo> [wiki] [[Brainfuck implementations]] https://esolangs.org/w/index.php?diff=52196&oldid=51890 * Copy * (-34) Update link from old website
03:05:17 -!- nullcone has joined.
03:12:49 <HackEgo> [wiki] [[User talk:Madk]] https://esolangs.org/w/index.php?diff=52197&oldid=18169 * MD XF * (+391)
03:13:08 <HackEgo> [wiki] [[User talk:Madk]] https://esolangs.org/w/index.php?diff=52198&oldid=52197 * MD XF * (-39)
03:13:26 <HackEgo> [wiki] [[User talk:Madk]] M https://esolangs.org/w/index.php?diff=52199&oldid=52198 * MD XF * (-1)
03:19:31 -!- Remavas has joined.
03:21:10 <lambdabot> CYVR 190200Z 09005KT 15SM FEW006 OVC016 16/14 A3017 RMK CF1SC7 SLP218
03:21:40 <zzo38> (There is another airport near where I am but which does not have METAR)
03:21:52 <zzo38> (I don't know why they don't have METAR.)
03:40:03 -!- Remavas has quit (Read error: Connection reset by peer).
03:41:07 -!- Remavas has joined.
03:45:20 -!- sleffy has joined.
03:51:05 -!- moony has joined.
04:03:06 -!- augur has joined.
04:03:54 <HackEgo> [wiki] [[Talk:Brainfork]] https://esolangs.org/w/index.php?diff=52200&oldid=41792 * CatIsFluffy * (+196) :P
04:13:43 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
04:24:45 <zzo38> Are you going to make Deadfish implementation with more kinds of assembly languages (other than just Famicom and MMIX)?
04:33:04 -!- moony has quit (Ping timeout: 255 seconds).
04:49:56 <Sgeo_> I have been getting a 24 Hour Server Suspension Warning every 24 hours or so for the past 3 days.
04:58:46 <zzo38> What is that warning for?
04:58:58 <oerjan> have you paid you Maintenance Fee (TM)?
04:59:10 -!- MDude has quit (Ping timeout: 255 seconds).
04:59:28 -!- MDude has joined.
05:14:51 -!- nullcone has quit (Quit: Connection closed for inactivity).
05:18:47 -!- augur has quit (Remote host closed the connection).
05:31:27 -!- jaboja has quit (Ping timeout: 255 seconds).
05:35:07 -!- augur has joined.
05:59:49 -!- erkin has quit (Quit: Leaving).
06:01:11 -!- augur has quit (Remote host closed the connection).
06:18:01 -!- augur has joined.
06:20:54 -!- augur has quit (Remote host closed the connection).
06:22:37 -!- augur has joined.
06:35:43 -!- sdhand has quit (Excess Flood).
06:35:52 -!- sdhand has joined.
06:36:06 -!- sdhand has quit (Changing host).
06:36:06 -!- sdhand has joined.
07:00:27 <Sgeo_> And no desire to, I haven't been using it at all iirc
07:03:12 -!- FreeFull has quit.
07:13:19 <oerjan> Sgeo_: well int-e closed his account with them after they announced that (badly).
07:13:29 -!- oerjan has quit (Quit: Nite).
07:51:59 -!- doesthiswork has quit (Quit: Leaving.).
07:57:02 -!- erkin has joined.
07:58:23 <zzo38> I made a list of some new kind of magic item that could be used in a GURPS game or other kind of game; these are the (mostly) rare kind so is not common to just find it in any shop, to make one, etc, but sometimes it can be.
07:58:55 <zzo38> It is: http://zzo38computer.org/textfile/miscellaneous/gurps/newitem
08:04:19 -!- jaboja has joined.
08:04:33 -!- nullcone has joined.
08:05:38 -!- Mayoi has joined.
08:06:01 -!- erkin has quit (Disconnected by services).
08:06:03 -!- Mayoi has changed nick to erkin.
08:17:14 -!- nullcone has quit.
08:24:15 <shachaf> oerjan: did you read emil books
08:31:39 -!- augur has quit (Remote host closed the connection).
08:58:02 -!- sleffy has quit (Ping timeout: 258 seconds).
09:04:59 -!- wob_jonas has joined.
09:06:00 -!- erkin has quit (Quit: Leaving).
09:06:49 <wob_jonas> int-e: it still looks like your attempts don't fit into an irc line, but it's shorter than what I tried
09:06:55 <wob_jonas> I'll have to check if it works and how it works
09:10:36 <wob_jonas> <zzo38> Is there a such things as MMIX JIT to compile a MMIX code into a code for a different computer? => um, maybe qemu has one? it has jits from multiple pairs of archs I think
09:25:10 -!- nullcone has joined.
09:32:45 <int-e> wob_jonas: the last one actually did make it to the bot in one piece. 451 characters is really pushing the limits though. see http://sprunge.us/cBXQ for the whole thing
09:33:24 <int-e> (and I have a short hostmask, which probably helps)
09:33:29 -!- jaboja has quit (Ping timeout: 268 seconds).
09:34:00 <wob_jonas> int-e: ah, so I should try from my b_jonas account or something
09:42:10 <HackEgo> [wiki] [[Integ]] N https://esolangs.org/w/index.php?oldid=52201 * B jonas * (+228) Created page with "'''Integ''' is a programming language by __kerbal__. Some documentation and an implementation at https://github.com/kerbin111/Integ [[Category:Languages]] Category:Impleme..."
09:43:30 <HackEgo> [wiki] [[Language list]] https://esolangs.org/w/index.php?diff=52202&oldid=52163 * B jonas * (+12)
09:50:16 -!- nullcone has quit.
09:51:28 <zzo38> For MMIX JIT, my idea involving only user programs anyways (not the operating system), and to take advantage of some of the features such as needing SYNCID for self-modifying code. Still is not quite the kind of code to be JIT, although is closer than some machine instruction sets.
09:54:38 <wob_jonas> sorry, it seems qemu doesn't yet have mmix emulation http://git.qemu.org/?p=qemu.git;a=tree;f=target;h=fc7882c64d1d9ded74202aeeff083fd3c5e52c46;hb=master
09:56:04 <wob_jonas> int-e: it looks like the new tricks you use is to decode digits through separate memory cells, and use @() to do this efficiently
09:56:26 <wob_jonas> (efficiently in source code space)
09:58:28 <wob_jonas> and it indeed looks shorter than the latest I have, which is http://dpaste.com/1C3K0R1.txt
10:01:24 <wob_jonas> also one more trick is to store the data to cell 0, which I really should have done
10:13:16 <int-e> I was really quite happy to have a good use for @ and _
10:20:17 <zzo38> Another use of LDHT instruction in MMIX is to be able to do multiple kind of endianness conversion (with MOR) by using only one constant, so you can read 16-bit small-endian, 32-bit small-endian, and 32-bit PDP-endian. When converting endianness of signed numbers, you can also use one constant (already mentioned in the MMIX documentation, I think), and follow by SR to tell how much.
10:29:24 -!- LKoen has joined.
10:46:48 -!- hkgit03 has joined.
11:33:34 -!- boily has joined.
12:01:16 <Taneb> :D I have a job offer!
12:02:48 <lambdabot> quintopia said 18h 51m 39s ago: helloily. i'm back. how's this coming saturday for ya?
12:04:21 <boily> @tell quintopia qunthellopia! can you make it Sunday? Saturday is the Fête Nationale and I'll be out.
12:05:36 -!- augur has joined.
12:06:41 <HackEgo> companion cube//There's cake inside it. Tear it apart, rip open your companion, and extract the delicious, delicious cake...
12:07:29 <Taneb> So I guess I'm moving to Cambridge
12:09:02 -!- MDude has quit (Ping timeout: 240 seconds).
12:09:14 <boily> what's your job offer?
12:09:42 <Taneb> 9 month contract working on https://www.myrtlesoftware.com/2017/04/20/deep-learning/
12:10:07 -!- augur has quit (Ping timeout: 240 seconds).
12:11:56 <boily> in fact, quite nice.
12:15:01 <Taneb> Yeah, I'm really excited by it!
12:16:50 -!- sdhand has changed nick to dnahds.
12:28:34 -!- boily has quit (Quit: TRUNK CHICKEN).
12:58:19 <olsner> shachaf: emil i lönneberga?
12:58:38 <olsner> never read the books, just saw the tv series (or perhaps the movie)
13:06:15 -!- augur has joined.
13:11:29 -!- augur has quit (Ping timeout: 246 seconds).
13:39:46 -!- MDude has joined.
14:00:26 -!- doesthiswork has joined.
14:07:31 -!- augur has joined.
14:12:23 -!- augur has quit (Ping timeout: 246 seconds).
14:20:14 -!- nullcone has joined.
14:32:18 -!- `^_^v has joined.
14:34:00 <wob_jonas> Taneb: nice! do you also have your university degree complete?
14:37:27 <wob_jonas> automating car driving? meh, I'll believe it only if people develop those sorts of things specifically targeting driving in bad weather and big cities in the uglier parts of Europe. it's easy to get better systems for driving on american highways.
14:39:24 <wob_jonas> I've worked enough with computer vision to know that every sort of automation is at least ten times more difficult if you want it to work in all times of the day and night in all lighting conditions and weather, not just nice bright light with a thin cloud in front of the sun to remove shadows.
14:42:17 -!- dnahds has quit (Excess Flood).
14:42:26 -!- sdhand has joined.
14:42:31 -!- sdhand has quit (Changing host).
14:42:31 -!- sdhand has joined.
15:08:18 -!- augur has joined.
15:11:15 <LKoen> Taneb: congratulations
15:12:56 -!- augur has quit (Ping timeout: 246 seconds).
15:12:57 -!- hkgit03 has quit (Quit: Leaving).
15:16:53 -!- jaboja has joined.
15:22:39 <HackEgo> [wiki] [[Special:Log/newusers]] create * Electron * New user account
15:30:02 -!- heroux has quit (Ping timeout: 240 seconds).
15:30:25 -!- heroux has joined.
15:36:41 <HackEgo> [wiki] [[Esolang:Introduce yourself]] https://esolangs.org/w/index.php?diff=52203&oldid=52195 * Electron * (+160)
15:40:07 -!- alercah has left.
15:40:25 <HackEgo> [wiki] [[Talk:Call Queue]] https://esolangs.org/w/index.php?diff=52204&oldid=43561 * Electron * (+206) /* tasq */ new section
15:56:48 -!- __kerbal___ has joined.
15:57:23 -!- __kerbal___ has quit (Client Quit).
16:09:04 -!- augur has joined.
16:11:55 -!- jaboja has quit (Ping timeout: 240 seconds).
16:14:16 -!- augur has quit (Ping timeout: 258 seconds).
16:15:17 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
16:37:48 -!- izabera has changed nick to offesa.
16:38:47 -!- offesa has changed nick to izabera.
16:42:26 <HackEgo> [wiki] [[FFM/FFB]] N https://esolangs.org/w/index.php?oldid=52205 * Enoua5 * (+6979) Created page with "{{infobox proglang |name=Foxrabbit's Finite-state Map/Foxrabbit's Finite-state Binary |paradigms=[[:Category:Declarative paradigm|Declarative]] |author=[[User:Enoua5]] |majori..."
16:44:10 <HackEgo> [wiki] [[FFM/FFB]] M https://esolangs.org/w/index.php?diff=52206&oldid=52205 * Enoua5 * (+2) /* Turing Completeness */
16:47:20 -!- enoua5 has joined.
16:47:37 -!- Remavas has quit (Ping timeout: 258 seconds).
16:47:55 <enoua5> BOOM! Made an entire language within a single day! Opinions? https://esolangs.org/wiki/FFM/FFB
16:51:40 <HackEgo> [wiki] [[FFM/FFB]] M https://esolangs.org/w/index.php?diff=52207&oldid=52206 * Enoua5 * (+155)
16:55:49 <HackEgo> [wiki] [[Language list]] M https://esolangs.org/w/index.php?diff=52208&oldid=52202 * Enoua5 * (+14) /* F */
16:56:37 <izabera> https://twitter.com/anne_theriault/status/876235898500861954
16:59:53 <HackEgo> [wiki] [[User:Enoua5]] M https://esolangs.org/w/index.php?diff=52209&oldid=52159 * Enoua5 * (+62) /* Languages created */
17:01:42 -!- `^_^v has quit (Quit: This computer has gone to sleep).
17:02:19 -!- enoua5 has left.
17:10:07 -!- augur has joined.
17:13:56 -!- GeekDude has joined.
17:14:47 -!- augur has quit (Ping timeout: 240 seconds).
17:24:40 <HackEgo> [wiki] [[Small]] https://esolangs.org/w/index.php?diff=52210&oldid=45780 * Get52 * (-36) User input was made deprecated.
17:26:31 <lambdabot> LOWI 191620Z 05011KT 020V080 9999 FEW080 30/15 Q1018 NOSIG
17:26:44 <int-e> hmm, maybe I won't walk home today
17:27:00 <HackEgo> [wiki] [[Small]] https://esolangs.org/w/index.php?diff=52211&oldid=52210 * Get52 * (+48)
17:27:44 <HackEgo> [wiki] [[Small]] https://esolangs.org/w/index.php?diff=52212&oldid=52211 * Get52 * (+2)
17:28:52 -!- MrBismuth has joined.
17:29:01 -!- MrBusiness has quit (Ping timeout: 258 seconds).
17:29:21 <\oren\> These "jeans" do not feel like denim at all
17:30:47 -!- FreeFull has joined.
17:39:02 <HackEgo> [wiki] [[FFM/FFB]] M https://esolangs.org/w/index.php?diff=52213&oldid=52207 * Enoua5 * (+7) fixed small bug in revcat
17:43:20 <shachaf> olsner: Ah. I think I read at least one of the books but it was a long time ago.
17:43:27 <shachaf> What about Emil and the Detectives?
17:44:09 <olsner> never heard of it, seems to be german
17:44:27 <HackEgo> [wiki] [[Talk:Brainfork]] https://esolangs.org/w/index.php?diff=52214&oldid=52200 * Zseri * (+121)
17:54:47 -!- `^_^v has joined.
17:57:44 -!- LKoen has quit (Remote host closed the connection).
18:09:35 <zzo38> Is there some kind of fast way to make Morton numbers using MMIX instructions?
18:10:55 -!- augur has joined.
18:10:58 <shachaf> Do you mean interleaving the bits of integers?
18:15:27 -!- augur has quit (Ping timeout: 240 seconds).
18:17:00 <int-e> I would expect the usual 0x55..., 0x33..., 0x0F... bisection
18:17:09 -!- moony has joined.
18:17:27 <int-e> (though not, actually, in that order)
18:21:01 <HackEgo> smlist 454: shachaf monqy elliott mnoqy Cale
18:22:38 <HackEgo> Non-update notification for the webcomic Super Mega.
18:24:29 <int-e> and of course the masks aren't quite the same: x = x & 0xFFFFUL | ((x & 0xFFFF0000UL) << 16); x = x & 0xFF000000FFUL | ((x & 0xFF000000FF00UL) << 8); x = x & 0xF000F000F000FUL | ((x & 0xF000F000F000F0UL) << 4); x = x & 0x303030303030303UL | ((x & 0xC0C0C0C0C0C0C0C0UL) << 2); x = x & 0x1111111111111111UL | ((x & 0x2222222222222222UL) << 1);
18:25:20 <int-e> (untested, and 0xC0C0C0C0C0C0C0C0UL should be 0xC0C0C0C0C0C0C0CUL, so that's at least one bug)
18:27:17 <int-e> well Germany has introduced the dual system. https://de.wikipedia.org/wiki/Duales_System_Deutschland
18:28:02 <shachaf> Did you read any Emil book?
18:28:05 <olsner> TODO: implement https://twitter.com/MarieLaureGB/status/874343740860772353 in self-modifying Piet
18:28:25 <int-e> the detective? children's books? I suppose I have, at some point.
18:29:07 <int-e> actually, no... children is the plural already... stupid inconsistent grammar.
18:29:20 <int-e> shachaf: but I don't see any connection.
18:31:42 -!- erkin has joined.
18:33:14 -!- augur has joined.
18:35:33 -!- wob_jonas has joined.
18:38:40 -!- wob_jonas has quit (Client Quit).
18:39:51 <int-e> and (looking at volume 4a), you can apparently play some tricks with M(X)OR.
18:42:04 <zzo38> Yes I also thought about using MOR and possibly MXOR
18:42:45 <HackEgo> [wiki] [[FFM/FFB]] M https://esolangs.org/w/index.php?diff=52215&oldid=52213 * Enoua5 * (+34)
18:42:53 <zzo38> Which might reduce how many of these masking you should need.
18:42:57 -!- wob_jonas has joined.
18:43:49 -!- Phantom_Hoover has joined.
18:43:58 <wob_jonas> zzo38: I think you can interleave bits quickly by using MOR a few times, on different sides
18:45:55 <zzo38> I know that you can reverse bits in a 64-bit number by using the constant #0102040810204080 once in the Y operand and once in the Z operand of a MOR instruction
18:46:02 <wob_jonas> anyway, you can move the bits around within any one byte using an MOR, in the same arrangement separately on 8 bytes
18:46:33 <wob_jonas> so you can use that to move either the lower or upper four bits of a byte to either the even or odd bits, or backwards
18:47:05 <wob_jonas> then you can use MOR the other way to rearrange bytes, OR to combine words, and even shifts can help
18:47:43 <wob_jonas> depending on what size of inputs and outputs you want, I think you can interleave bits quickly using that
18:48:22 <wob_jonas> zzo38: how big are your input numbers, and do you want this on a single pair of numbers or multiple together?
18:48:48 <zzo38> I should assume two 32-bit numbers.
18:49:40 <wob_jonas> if you only have one pair, then I think this is one of the many situations where x86's various random instructions on big vectors help you more than MMIX's MOR and stuff
18:51:47 -!- jaboja has joined.
18:52:28 <int-e> 3 MORs, two ANDs, one OR, will give you the "spread" (interleaving with 0 bits).
18:59:10 -!- moony has quit (Ping timeout: 260 seconds).
18:59:14 <wob_jonas> let me try to figure this out exactly
19:03:09 <int-e> here is a sketch using 4x4 matrices: http://sprunge.us/CMIF
19:03:41 <int-e> (I'm not claiming that it's optimal, but it does show that 3 MORs, 2 ANDs and one OR suffice)
19:04:57 <int-e> and I may actually be able to do the whole interleaving in the same number of operations? HMM.
19:06:26 -!- sleffy has joined.
19:08:41 <shachaf> are you an evil financier now
19:12:47 -!- [io] has joined.
19:12:47 -!- [io] has quit (Changing host).
19:12:47 -!- [io] has joined.
19:12:58 <shachaf> not yet evil or not yet financier
19:13:02 <shachaf> what sorts of things are you training to do
19:15:41 -!- betaveros_ has joined.
19:15:50 <Phantom_Hoover> other than 'writing code for kdb' i don't exactly know
19:18:42 -!- doesthiswork1 has joined.
19:18:56 -!- rodgort` has joined.
19:19:37 -!- doesthiswork has quit (*.net *.split).
19:19:37 -!- rodgort has quit (*.net *.split).
19:19:37 -!- betaveros has quit (*.net *.split).
19:19:37 -!- sparr has quit (*.net *.split).
19:19:37 -!- Cale has quit (*.net *.split).
19:19:37 -!- iovoid has quit (*.net *.split).
19:22:30 -!- j-bot has quit (Ping timeout: 260 seconds).
19:25:36 -!- sparr has joined.
19:25:36 -!- Cale has joined.
19:33:51 -!- Phantom_Hoover has quit (Ping timeout: 255 seconds).
19:34:45 <wob_jonas> zzo38: I think something like this should work, but it's untested http://dpaste.com/0ZP3BS8
19:35:07 <int-e> zzo38: http://sprunge.us/FXMP ... it appears that 4 MORs, 2 ANDs, 2 ORs are enough to do the interleave operation (plus a shift and another or to get both parts into the same register)
19:35:24 <wob_jonas> this uses a combination of 8 MOR plus 3 OR instructions, plus a few instructions that create constants and so are needed only once
19:37:13 <wob_jonas> i'm not saying this is optimal, it may be possible to do one or two shorter
19:39:22 <wob_jonas> it's much easier on future x86 cpus, which will have bit-collect (intercal select) and bit-uncollect instructions, so you'll be able to do this in just two bit-uncollect then an OR instruction
19:39:40 <int-e> (or three or four)
19:42:23 <int-e> there's an open question whether MXOR could make this even more efficient, but I don't really want to think about that. linear algebra is hard.
19:48:54 <int-e> (why did I say 2 ORs when I meant 1 OR)
19:51:34 <wob_jonas> (it's not really intercal select, mind you, just half of it)
19:54:02 <wob_jonas> I have an idea for an esoteric programming language. I'll have to figure out what exactly the rules I'm thinking of imply, and if there's a right set of features to keep to get something that is not completely unusable but also not trivially easy.
19:55:00 -!- jaboja has quit (Ping timeout: 255 seconds).
20:10:57 -!- hppavilion[1] has joined.
20:11:55 -!- hppavilion[1] has quit (Client Quit).
20:12:12 -!- hppavilion[1] has joined.
20:58:23 <fizzie> "Future", "will have"? Aren't those what PDEP and PEXT already are?
20:58:54 <wob_jonas> fizzie: yes, it's PDEP and PEXT, but most of the cpus you use right now don't yet support those
21:02:47 <fizzie> Aren't they just part of BMI2, in Haswell and later?
21:03:31 <int-e> . o O ( Flags: ... bmi2 ... )
21:03:39 <wob_jonas> fizzie: no, I think they're part of BMI3
21:03:55 <int-e> it seems that you're wrong
21:04:07 -!- Remavas-Hex has joined.
21:04:36 <shachaf> Netcraft confirms it: Intel is dying
21:04:56 <int-e> ARM SoCs are winning, I suppose
21:06:00 <fizzie> https://software.intel.com/sites/landingpage/IntrinsicsGuide/ says cpuid flags BMI2 for the pdep intrinsic.
21:06:36 <fizzie> And lists Haswell/Broadwell/Skylake perf numbers for it.
21:06:56 <wob_jonas> shachaf: I don't think they're dying. they're using dirty tricks successfully to remain the market leader. but AMD is putting up a really good fight: they've finally started making CPUs in all categories from middle-strength desktop to high-end server that are better than Intel's again, and they bought a graphics card manufacturer
21:07:10 <wob_jonas> fizzie: wtf, what does BMI3 do then?
21:08:09 <int-e> what is that bmi3 you're talking about
21:09:42 <wob_jonas> uh... I'm not at home, let me re-download the manuals to here
21:09:45 <int-e> (google finds ABM, BMI1, BMI2, TBM, but I have not seen any other bit manipulation stuff.
21:13:43 <wob_jonas> the doc says PDEP is in the BMI2 extension set, yes
21:13:56 <wob_jonas> let me look up what cpus have that
21:14:54 <fizzie> Flaky hotel WiFi here as well, so won't be around. I just remember being excited about having an opcode matching an intercal instruction so well. Doesn't look like people have been that interested about it in general.
21:16:00 <wob_jonas> fizzie: it is interesting, but sort of drowned out by all the other more important good and bad things about AVX512
21:18:35 <wob_jonas> fizzie: ok, you're right, it's probably not just in future cpus, but it's still not in cheaper cpus you buy today
21:20:36 <FireFly> PDEP/PEXT are interesting, makes me want to write a compiler of some kind for an APL-like
21:20:59 -!- mroman has joined.
21:23:57 <HackEgo> [wiki] [[Special:Log/newusers]] create * Lealcy * New user account
21:26:09 -!- jaboja has joined.
21:29:25 <mroman> 9827568326>\#+:#*9-#\_$.@
21:30:46 <int-e> doesn't look like much
21:32:36 <wob_jonas> but even without PDEP, you can do the spreading of an integer to every other bits with like five AVX2 instructions I think
21:33:49 <wob_jonas> hmm no, maybe it's more because AVX2 doesn't have a full shuffle
21:36:15 -!- LKoen has joined.
21:40:05 <wob_jonas> ok, make that six instructions then
21:43:11 <wob_jonas> anyway, you could probably do the full interleave in like ten instructions
21:44:17 <wob_jonas> I dreamed up an esoteric language, and I'm not sure if it's literally turing-complete
21:45:19 <LKoen> turingally leteracy-complete
21:45:42 <mroman> the name is turing complete?
21:46:13 <wob_jonas> it is quite powerful computationally, so in practice you could do any computation with it, but I don't know whether it can use arbitrarily large amount of memory completely (as opposed to just access a memory that grows at least tetration quickly in the program size)
21:56:00 <mroman> now I have to write a password cracker that understands befunge
21:56:09 <mroman> to crack my own password of course
21:56:25 <mroman> the new captchas are befunge code
21:56:55 <wob_jonas> oh. but aren't there only a few of them?
21:57:09 -!- AnotherTest has joined.
21:57:18 <mroman> not if you have like 10 passwords candidates in your head you might have used for this account.
21:57:25 <mroman> then it's really annoying.
21:57:48 <mroman> and all of these password candidates either have a 0,1,2,3,4,5,6,7,8,9 as a last digit
21:57:52 <wob_jonas> no, I mean only like 10 different befunge captchas
21:57:57 <mroman> then you're easily at 100 passwords you gotta try
21:58:10 <wob_jonas> that you have to solve once, and then you always get only those ten for each attempt
21:58:35 <mroman> this is like hashcash captchas
21:58:42 <mroman> you colud have hashcash captchas
21:58:49 <mroman> that take 1min to solve on a modern computer :D
21:59:54 <wob_jonas> is that useful? aren't they the kind of hashes where you can get the page a million times, solve the million hashes in parallel on a GPU in two minutes, then send a million password attempts?
21:59:59 <mroman> but having to interpret code is even better
22:00:03 <mroman> some programs might not terminate
22:00:06 <mroman> that oughta show them cracker
22:00:33 <mroman> if you can solve a million hashes on a GPU in two minutes
22:00:42 <mroman> then you should pick another kind of hash
22:00:56 <mroman> or increase the amounts of bits
22:00:57 <wob_jonas> maybe only a hundred thousand, not an actual million
22:01:16 <mroman> i.e. instead of requiring to find a hash matching the first 16 bits of this hash
22:01:23 <mroman> you require it to match the first 32bits or something
22:01:47 <wob_jonas> but then it really depends on what computer the user has
22:02:08 <wob_jonas> some computers would take 30 minutes, others only 1 minute on the same challenge
22:02:16 <mroman> the whole point is to waste cpu cycles of an attacker
22:02:35 <mroman> and increase global warming :D
22:03:51 <wob_jonas> isn't it better to use human captchas and increase employment?
22:05:03 <wob_jonas> you know, captchas only humans can solve
22:05:20 <mroman> I've been thinking about a new kind of captchas
22:05:34 <mroman> it'll display you N sentences of a conversation out of order
22:05:38 <mroman> and you have to order them
22:05:58 <wob_jonas> nah, that usually doesn't have a unique solution
22:06:26 <wob_jonas> not in conversations you can easily mass-generate
22:06:28 <mroman> they do the same kind of tests to test for autism and stuff
22:06:35 <mroman> so one problem is probably that you exclude autists
22:07:15 <mroman> actually they do it with pictures of people doing things and you have to order the pictures to form a coherent story
22:07:47 <wob_jonas> just use some known hard problem like driving cars at night in a city. you have to remote-drive the car using live camera feeds, and you manage to drive it to the destination (which is about a minute away from the source) without crashing, you succeeded the captcha. if you have a high latency net connection, you can't complete this sort of captcha.
22:08:01 <wob_jonas> mroman: those are even more difficult to mass-produce
22:08:02 <mroman> that was a waste of 3000 bucks
22:08:56 <mroman> the test for autism I took 2 years ago
22:09:29 <mroman> it would produce a new job sector
22:09:34 <wob_jonas> oh, like the policeman and the IQ test?
22:09:47 <mroman> gotta create new jobs for those coal miners
22:10:48 -!- AnotherTest has quit (Quit: ZNC - http://znc.in).
22:10:56 <wob_jonas> policeman joke. "I was on an IQ test yesterday." "Did you get the result yet?" "<happy> Yes, it's negative."
22:11:37 <mroman> Asperger people aren't stupid
22:11:54 <mroman> It's not that hard to know when something is supposed to be funny
22:12:32 <mroman> these tests measure more your social skills
22:12:48 <mroman> they measure the _pure_ basics of your social skills
22:13:02 <mroman> like showing you faces with specific facial expressions
22:13:10 <mroman> and then assign emotions to the facial expression
22:13:49 <mroman> even if you have very poor social skills you can still tell when a person is smiling or crying
22:13:56 <mroman> those are _very_ obvious facial expressions
22:14:24 <mroman> I had a lot of tests done
22:14:33 <mroman> well health insurance paid it
22:14:42 <wob_jonas> did they at least do useful tests too?
22:14:51 <mroman> you mean like rorschach test?
22:14:54 <mroman> they did those too yes
22:15:56 <wob_jonas> ones that made you feel it was worth paying money for
22:15:57 <mroman> they test wether you can recognize faces of famous people, the test wether you can assign emotions to facial expressions
22:16:05 <mroman> they test wether you can order these picture stories
22:16:15 <mroman> they test your memory by having you remember 10 words
22:16:38 <wob_jonas> I can recognize some of them, but not many
22:16:51 <mroman> they test if your concentration with these p flipped-p test where you have a piece of paper with hundreds of p's some of them are flipped/mirrored and
22:17:02 <mroman> you have to circle the flipped ones as fast as you can.
22:17:19 <wob_jonas> so are they all psychological tests?
22:17:42 <mroman> they have a board with 6 circles and they touch them in some order and you have to repeat that order
22:17:54 <mroman> and then a lot of the "weird" tests such as Rorschach
22:18:07 <mroman> and "build patterns with these colored squares"
22:18:15 <mroman> those are probably the unscientific tests
22:18:20 <mroman> the other ones are at least rooted in some science
22:18:26 <wob_jonas> draw the current time on an analog clock face?
22:18:36 <mroman> no, that I didn't have to do
22:18:44 <mroman> that sounds more like something you do with dementia patients
22:19:26 -!- Phantom_Hoover has joined.
22:19:35 <mroman> apparentely I'm unable to form logical conclusions
22:20:02 <wob_jonas> "unable to form logical conclusions"?
22:20:16 <mroman> that sounds like completely made up
22:20:19 <mroman> I teach computer science
22:20:29 <mroman> it would be pretty bad if I couldn't form logical conclusions.
22:20:48 <wob_jonas> yes, although there are a few very horrible teachers out ther
22:21:10 -!- wob_jonas has quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client).
22:21:38 -!- Lymia has quit (*.net *.split).
22:21:38 -!- dingbat has quit (*.net *.split).
22:21:39 -!- shikhin has quit (*.net *.split).
22:21:39 -!- myname has quit (*.net *.split).
22:21:46 -!- mynery has joined.
22:21:51 -!- shikhin has joined.
22:22:17 -!- Lymia has joined.
22:22:20 <mroman> If it rains streets are wet. If streets are wet, did it rain?
22:22:24 -!- dingbat has joined.
22:22:41 <mroman> Those are questions you can use to establish if someone can think logically
22:22:51 <mroman> so I have no idea where the fuck this comes from.
22:23:11 <LKoen> mroman: do you know the tv series "orphan black"?
22:23:32 <LKoen> there are quite a few episodes in which some of the characters have to submit to daily psychological tests
22:24:09 <LKoen> and thoses tests consists mainly in those kind of questions
22:24:13 -!- DHeadshot has joined.
22:24:17 <mroman> I think they can't tell the difference between what's possible and what's not possible.
22:25:02 <LKoen> they're given a syllogism, and have to answer yes or no depending on whether the conclusion is consistent with the two priors
22:25:18 <LKoen> sometimes the conclusion is false, because the priors are false, but that's still a yes if that's consistent
22:25:40 -!- Remavas-Hex has quit (Quit: Reconnecting, hold on).
22:25:40 <mroman> those questions are btw also good for turing tests :D
22:26:27 <LKoen> Taneb: can you train a neural network to answer those questions?
22:26:48 <mroman> One of my useless "talents" is to overthink a lot.
22:27:06 -!- Remavas-Hex has joined.
22:27:20 <mroman> i.e. there's the common believe that if you voluntarily admit yourself into a mental hospital you can leave whenever you want.
22:27:36 <mroman> but it's objectively wrong.
22:27:46 <mroman> there's no such clause/law that actually says this.
22:29:25 -!- jaboja has quit (Ping timeout: 240 seconds).
22:29:47 <mroman> so the logical conclusion is that even if you volunarily admit yourself you're actually not guaranteed to be able to get out again.
22:30:20 <mroman> but apparentely psychologists don't accept this as a logical conclusion.
22:34:13 -!- `^_^v has quit (Quit: This computer has gone to sleep).
22:34:51 <mroman> There really is no such law that specifies this.
22:35:10 <Phantom_Hoover> you should read up on paraconsistent logics or something
22:36:00 -!- Sgeo_ has quit (Read error: Connection reset by peer).
22:38:17 <mroman> inconsistency tolerant logic
22:39:20 <mroman> "Dialetheism is the view that some statements can be both true and false simultaneously."
22:40:24 <mroman> > Ambiguous situations may cause humans to affirm both a proposition and its negation. For example, if John stands in the doorway to a room, it may seem reasonable both to affirm that John is in the room and to affirm that John is not in the room.
22:40:26 <lambdabot> <hint>:1:97: error: parse error on input ‘,’
22:40:44 <mroman> that's just a matter of definition though.
22:40:56 -!- Remavas-Hex has quit (Remote host closed the connection).
22:41:04 <mroman> the result isn't of a true contradiciton it's because a term is not well-defined enough for everybody to agree on it
22:41:13 <LKoen> John \cap Room \neq {}
22:41:30 <LKoen> and John \cap cRoom \neq {}
22:41:39 <mroman> if you'd define what exactly is "in the room" then you will have a clear true/false answer
22:41:40 <LKoen> not inconsistent at all :)
22:41:56 <LKoen> I think the ambiguity came from John being larger than the doorstep
22:42:06 <LKoen> just like the Alps are both in France and in Italy
22:45:49 <mroman> I refuse statements that are both true and false :D
22:47:05 <LKoen> it's jsut that the statement make it sound like John is a point
22:47:08 <LKoen> when he's clearly a fat set
22:48:21 -!- nullcone has quit.
22:50:32 -!- mroman has quit (Ping timeout: 260 seconds).
22:54:32 -!- mynery has changed nick to myname.
22:56:11 -!- boily has joined.
22:56:16 <HackEgo> porthello//Hellonfused one. Porthellos are the standard greeting format in #esoteric. Best enjoyed with some thé or caffè and a fternooner.
22:56:20 <lambdabot> quintopia said 6h 44m 33s ago: sure. what time?
22:56:46 <boily> @ask requinthellopia. about 8pm?
22:57:01 <LKoen> is caffè a machine learning ide?
23:00:37 <boily> `cwlprits porthello
23:01:17 <HackEgo> 8751:2016-07-05 <oerjän> le/rn porthello/Hellonfused one. Porthellos are the standard greeting format in #esoteric. Best enjoyed with some th\xc3\xa9 or caff\xc3\xa8 and a fternooner.
23:01:19 <boily> I *think* I mentioned caffè in thère because it has an è in it, to counteract the thé's é.
23:04:41 -!- LKoen has quit (Quit: “It’s only logical. First you learn to talk, then you learn to think. Too bad it’s not the other way round.”).
23:15:10 -!- staffehn has quit (Ping timeout: 258 seconds).
23:15:10 <HackEgo> [wiki] [[Esolang:Introduce yourself]] M https://esolangs.org/w/index.php?diff=52216&oldid=52203 * Lealcy * (+163)
23:15:39 <HackEgo> [wiki] [[PNID]] N https://esolangs.org/w/index.php?oldid=52217 * Lealcy * (+6287) Initial edit!
23:17:01 <HackEgo> [wiki] [[Language list]] M https://esolangs.org/w/index.php?diff=52218&oldid=52208 * Lealcy * (+11) Added ''PNID'' to the list of languages.
23:19:19 <HackEgo> [wiki] [[PNID]] M https://esolangs.org/w/index.php?diff=52219&oldid=52217 * Lealcy * (+11) /* Interpreter */
23:20:56 -!- staffehn has joined.
23:22:52 <\oren\> https://youtu.be/N6jVr9cTed4?t=1h
23:23:19 <HackEgo> [wiki] [[PNID]] M https://esolangs.org/w/index.php?diff=52220&oldid=52219 * Lealcy * (-1) /* Language overview */
23:39:50 -!- Sgeo has joined.
23:44:15 <HackEgo> goofix//Goofix is an antropomorphic canine arithmetic notation.
23:46:34 -!- MrBismuth has changed nick to MrBusiness.
23:52:11 -!- boily has quit (Quit: ORPHAN CHICKEN).
23:53:40 -!- DHeadshot has quit (Ping timeout: 258 seconds).
23:59:30 -!- oerjan has joined.