00:00:00 --- log: started forth/04.08.29 00:00:19 lol 00:00:26 rm -rf /tmp/* ? 00:02:55 NO 00:03:05 i did that and all my stickie notes disappeared :( 00:03:15 i wouldn't want to subject you to such trauma 00:03:43 'dd if=/dev/zero of=/dev/hda' will clean up *everything* :-) 00:05:15 lol 00:05:24 * I440r_ doesnt use sticky notes 00:05:33 if they were sticky i might be able to remember them 00:06:22 --- join: kc5tja (~kc5tja@66-74-218-202.san.rr.com) joined #forth 00:06:37 Hi kc5tja 00:06:37 --- mode: ChanServ set +o kc5tja 00:06:38 kc5! 00:06:40 Hey 00:06:47 Can't stay long. Have to get to bed soon. 00:06:59 im almost zoned out as it is lol 00:07:42 * kc5tja just got off of work, so I'm only kinda sorta zoned. 00:08:40 lol been there, done that 00:09:36 --- quit: madwork (Read error: 54 (Connection reset by peer)) 00:09:59 well if i got to sleep now i might be able to get isforht-fbsd running tomorrow 00:10:10 if i can figur out the weird shit thats going on 00:10:34 * kc5tja is trying to get FTS/Forth to parse words now. 00:10:44 cool 00:10:44 I worked it out while on work -- just basically plugging in source code now. 00:11:19 I'll try and work on dictionary support tomorrow if the work day is as slow as it was today. 00:12:46 well nite folks 00:12:53 im half dead 00:13:00 Good night I440r 00:13:06 only had 10 hours of sleep last nite 00:13:07 lol 00:13:43 night 00:13:51 Ahhh, poor baby. 00:13:56 I get, maybe, six. At most. 00:38:33 --- join: qFox (C00K13S@82-169-140-229-mx.xdsl.tiscali.nl) joined #forth 00:39:54 --- join: mur__ (~mur@kyberias.uiah.fi) joined #forth 00:40:15 --- quit: Sonarman ("Lost terminal") 00:45:06 --- quit: kc5tja ("leaving") 00:51:09 --- quit: mur_ (Read error: 110 (Connection timed out)) 01:25:29 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 01:34:58 --- quit: mur__ (Read error: 110 (Connection timed out)) 01:37:01 --- quit: scope (Client Quit) 02:08:27 --- quit: crc (Client Quit) 03:32:15 --- join: crc (crc@23-pool1.ras11.nynyc-t.alerondial.net) joined #forth 04:40:07 --- quit: crc ("Time for bed... Goodnight All!") 05:22:40 --- join: FlamingRain (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 05:46:20 --- quit: FlamingRain ("Leaving") 06:06:19 --- quit: I440r (Read error: 110 (Connection timed out)) 06:08:14 --- quit: mur_ (Remote closed the connection) 06:09:04 --- join: mur (~mur@kyberias.uiah.fi) joined #forth 06:30:47 * Tomasu is away: nigit 06:57:53 --- quit: slava (Read error: 110 (Connection timed out)) 07:00:42 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 07:10:58 --- quit: mur (Read error: 110 (Connection timed out)) 07:29:00 --- join: poing (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 07:36:43 --- quit: poing ("leaving") 07:36:57 --- join: poing_ (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 07:39:09 --- nick: poing_ -> zruty 07:42:02 grr slavas server is offline :( 07:43:09 got tired of ghosting other poings, so i switched nicks 07:44:00 k 07:48:16 guess i could have set kill on, but i cant really lay that much claim on the nick 07:48:48 methinks i'll set it for this one 07:49:07 darn, it's turned off on the network 08:05:29 --- quit: mur_ (Remote closed the connection) 08:06:03 --- join: mur (~mur@uiah.fi) joined #forth 08:35:28 --- join: Herkamire (~jason@h000094d30ba2.ne.client2.attbi.com) joined #forth 08:35:28 --- mode: ChanServ set +o Herkamire 08:39:32 just set kill on 08:39:34 :) 08:43:28 kill is off on the whole network 08:44:07 * zruty kinda likes zruty better, more distinctive 08:44:44 oh is it 08:44:54 shouldnt be in the help at all then 08:45:16 standard nickserv docs 08:45:26 no matter 08:45:27 :) 08:45:36 should definitely be customized for freenode tho 08:56:42 --- quit: zruty ("leaving") 09:16:29 --- join: zruty (~chuck@64-121-15-14.c3-0.sfrn-ubr8.sfrn.ca.cable.rcn.com) joined #forth 09:40:13 --- join: FlamingRain (~Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 09:48:39 --- quit: FlamingRain ("Leaving") 09:55:04 --- join: kc5tja (~kc5tja@66.74.218.202) joined #forth 09:55:11 --- mode: ChanServ set +o kc5tja 09:56:06 --- join: crc (crc@83-pool1.ras11.nynyc-t.alerondial.net) joined #forth 09:56:56 Greetings. Can't stay for very long at all. 09:57:04 Hi 09:57:20 Hello 10:00:58 * crc has finally release an RetroForth 7.3 (alpha) 10:01:54 Nice. 10:02:06 * kc5tja is still trying to figure out why his ParseWord word isn't working. 10:10:31 --- join: crc_ (crc@75-pool1.ras11.nynyc-t.alerondial.net) joined #forth 10:13:48 --- join: FlamingRain (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 10:15:37 Does **ANYONE** know how to get X11 tunneling working under SSH? 10:15:39 * kc5tja sighs 10:15:43 This is fucking rediculous! 10:17:50 At least I have shell access to my box from remote now. At least, until RR decides to change my box's IP address. Grrr! 10:18:37 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 10:20:43 slava: Any ideas on how to get SSH to port-forward X11? 10:20:55 I've got EVERY relavent option enabled, and it STILL won't fucking work. >:( 10:21:10 Internet HOW-TOs aren't any more enlightening. >:( 10:28:46 --- join: warpzero (~warpzero@dsl.103.mt.onewest.net) joined #forth 10:31:57 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 10:32:36 --- quit: crc (Read error: 110 (Connection timed out)) 10:42:52 --- quit: mur (Read error: 110 (Connection timed out)) 10:44:52 --- nick: crc_ -> crc 10:47:33 --- quit: kc5tja ("leaving") 11:02:52 --- quit: FlamingRain ("Leaving") 11:13:01 --- quit: crc ("Time for bed... Goodnight All!") 11:34:56 --- nick: I440r_ -> I440r 11:35:10 --- mode: ChanServ set +o I440r 11:58:50 --- join: mur (~mur@uiah.fi) joined #forth 12:06:16 --- join: FlamingRain (~Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 12:10:05 --- quit: mur_ (Read error: 110 (Connection timed out)) 12:16:14 hi 12:26:44 --- quit: slava (Read error: 104 (Connection reset by peer)) 12:28:42 --- quit: FlamingRain ("Leaving") 12:32:52 --- join: slava (~slava@CPE00096ba44261-CM000e5cdfda14.cpe.net.cable.rogers.com) joined #forth 12:33:44 --- join: mur_ (~mur@kyberias.uiah.fi) joined #forth 12:34:31 slava btw im about " <-- that close to rm -rf isforth-fbsd 12:34:35 :( 12:34:40 its pissing me off BIG fucking time 12:35:17 why? 12:36:36 because i cant fucking figure out why a syscall works ONE FUCKING TIME and then never works again 12:36:41 sysbrk 12:36:42 extend 12:36:43 save 12:36:46 run extended shit 12:36:51 the brk gives enomem 12:37:14 its the SAME FUCKING CODE!!!!!!!!!!!!!!! 12:38:19 run gdb kernel.com 12:38:24 break init_mem 12:38:25 run 12:38:30 single stel 3 instructions 12:38:39 (make note of address that the brk is set on) 12:38:54 quit out 12:39:03 run gdb isforth 12:39:09 break *0x804abf9 12:39:10 run 12:39:20 single step thru the int 0x80 and see e no fucking mem 12:40:25 have you asked #freebsd? 12:40:47 dont even go there 12:40:52 they dont know shit about it 12:40:59 ive emailed hackers@freebsd.org 12:41:05 ive tried everything 12:41:16 its just one fucking hurdle after the next 12:41:25 have you looked at how libc does brk? 12:41:27 it took me ONE fucking month to get this running in linux 12:41:29 PERIOD 12:41:35 err. libc is no help 12:41:51 why? 12:41:54 C is just portable assembler 12:42:02 libc sources are about as readable as gobbldegook 12:42:13 no it isnt 12:42:28 thats such a fucking crock of bullshit 12:42:37 c is write only 12:42:52 ive looked at the libc sources 12:42:56 ive looked at the kernel sources 12:43:03 ive looked in syscall.master 12:43:10 ive read man pages 12:43:23 if the code works in c though... 12:43:26 --- join: SDO (~SDO@67-23-111-213.clspco.adelphia.net) joined #forth 12:43:26 fucking man pages are "lets not REALLY tell them how this works" bullshit 12:43:38 I440r: that's what source is for 12:43:38 man pages are kind of useless i agree 12:43:44 zruty, unwinding a gordian knot is NOT my idea of fun 12:43:45 freebsd's source is lovely to behold 12:43:57 write a small program in c that does nothing but an isatty 12:44:07 single step it from the start to the syscall 12:44:15 it will take you about four years of single stepping 12:44:33 no fuck 12:44:35 but it will take the cpu a microsecond :) 12:44:42 well the tty handling stuff is a long and windy interface 12:44:45 thats like all the fucking morons who use cvs as a distrobution method 12:44:56 fucking cvs isnt a distrobution method and sources are not fucking documentation 12:45:07 you could trace it with something like valgrind perhaps 12:45:17 zruty, valgrind is useless 12:45:23 I440r, just relax, take a break, and come back to it later. :) 12:45:53 sure. instead of jsut giving me a fucking document describing the fucking syscall you write another fucking obfuscated bullshit utility to that i first have to learn and then have to interpret' 12:45:55 fuck it 12:46:00 slava 12:46:01 i have 12:46:03 and again 12:46:03 and again 12:46:04 and again 12:46:04 and again 12:46:05 and again 12:46:06 and again 12:46:08 --- quit: mur (Read error: 110 (Connection timed out)) 12:46:11 and again 12:46:13 and again 12:46:19 * zruty .oO( he's steamed ) 12:46:42 zruty, valgrind is only useful for debugging malloc/free 12:46:48 it doesn't do anything else 12:46:52 i thought it could trace all calls 12:47:30 not too familiar with it so i may just not know what i'm talking about 12:47:34 wouldn't be anything new :) 12:47:50 I440r, i've had many of those frustrating moments, don't worry 12:50:56 zruty, try debugging a garbage collector that uses tagged pointers. gdb is useless in this case 12:51:22 zruty, i had my to sprinkle code with printf's, and sift through megabytes of logging output :) 12:51:45 debugging a gc does not sound like much fun 12:52:01 sounds like you'd want the DEADBEEF approach 12:52:16 fill your memory with a pattern like DEADBEEF or something distinctive 12:52:30 eyeball hex dumps 12:53:07 recently i switched to non blocking i/o and printf didn't work anymore 12:53:13 that was hell 12:53:26 but eventually i figured out to send logging output to fd 3 :) 12:53:29 --- nick: zruty -> zruty|LUNCH 12:57:22 --- nick: zruty|LUNCH -> zruty 12:58:02 --- join: FlamingRain (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 13:22:32 --- quit: FlamingRain (Connection timed out) 14:18:12 --- quit: mur_ (Remote closed the connection) 14:18:44 --- join: mur (~mur@uiah.fi) joined #forth 14:19:12 hehe: http://ars.userfriendly.org/cartoons/?id=20040811&mode=classic 14:22:48 --- join: FlamingRain (Ecoder@c-24-129-95-254.se.client2.attbi.com) joined #forth 14:29:02 wished i got $500 :\ 14:29:55 i only got xbox, and thats not even half 14:32:58 --- join: TheBlueWizard (TheBlueWiz@pc9adn1d.ppp.fcc.net) joined #forth 14:32:58 --- mode: ChanServ set +o TheBlueWizard 14:33:13 ok i guess i kinda found a downside to forth. concatting strings is a pain in the ass 14:33:31 or actually, writing a concatted string 14:33:38 with variable length 14:33:41 "hello " "world" cat2 . 14:33:45 "hello world" 14:33:52 oh sure 14:33:55 :) 14:34:16 qFox: I'm sure there's plenty of forth code out there to do that 14:34:34 insert into 3x3moves ( name,move1,move2,move3,move4,move5,move6,move7,move8,move9,result,1wins,2wins,draws ) select 1,1,0,0,0,0,0,0,0,0,0,SUM(IF(result=1,1,0)),SUM(IF(result=2,1,0)),SUM(IF(result=3,1,0)) from 3x3 where move1 = 1 ; 14:34:48 ? 14:34:50 thats the query i have to write, over and over again 14:35:20 except the underlined stuff changes (or is sometimes not there, depending on how "Deep" the code is at that moment) 14:35:35 what are you trying to do??? 14:35:44 the code i've got right now, is uuuuugly and probably slow. 14:35:47 * Tomasu is back (gone 08:05:01) 14:35:58 thing is, this has to be done ... oh say 200milion times 14:36:15 and i dont think there is a fast way. 14:36:25 TheBlueWizard> computing, basically 14:37:25 it is too vague....I still have no idea what you are trying to do :P 14:37:56 the code creates all the moves, puts this data in a text file with fields etc, then creates a mysql file that first imports the other file into a certain table, then starts to use mysql to crunch the final data (in this case about 200milion records) 14:38:07 solve "connect4" game 14:38:23 strings are suxor. 14:38:39 and the way i'm doing it now... heh this just cant be fast :) 14:39:04 strings are a pain without gc 14:39:06 oh? hmm.... 14:39:22 anyways 14:39:36 the query for the second part is what i pasted above here, or an example of it 14:39:46 half of it is static, the other is variable 14:43:30 also, its a pain to write numbers 14:44:25 qFox: what do you mean solve? 14:44:34 well, of course you can use <# #> or something, but i think that's far too slow, knowing that it can only be 1 or 2 digits 14:44:38 trying to find the best first move or something? 14:44:42 connect4 14:44:52 no, well partially 14:44:55 qFox, why are you worried about performance? 14:44:58 write an AI to play against? 14:45:01 qFox, are you doing this billions of times over in a loop? 14:45:19 one part of the project is to solve the game, where you can force a win (it has already been proven but whatever), and find all possible ways to force a win 14:45:39 oh 14:45:55 the second is to compute all results and create an ai that merely picks his next move based on the move that can win the most 14:46:11 what shape is the board? 6x6? 14:46:19 heck no. 14:46:33 5x5? 14:46:36 a 6x6 board has a ... 25 digit number of moves 14:46:38 or something 14:46:48 no it doesn't 14:46:50 it has 6 14:46:51 right now i'm doing 4x4, which really is the max 14:46:53 no herk 14:47:03 total possible ways of playing 14:47:34 yes there are 6 slots, but there are about (width*height)! / (height!)^width ways to play the game 14:47:55 that number minus the "unreal" ways (moves after a win) 14:48:11 you drop a piece in, and it slides down 14:48:14 \ 4x4: Moves: 191448940 Games: 63063000 14:48:15 \ Moves: 97552623 Games: 33937264 14:48:15 \ Wins: 10536772 Draws: 23400492 14:48:15 \ Player 1 wins: 5303104 Player 2 wins: 5233668 14:48:18 there are never more than 6 moves 14:48:27 unless you're talking about a different game 14:48:33 no you are right 14:48:41 but 14:48:47 ok, lets take a 2x2 board 14:49:12 first you put in slot 1, then player 2 uses slot 1, player 1 uses slot 2, player 2 uses slot 2 14:49:19 board is filled, draw is reached 14:49:23 thats ONE game. 14:49:27 with 4 moves 14:49:28 are you guys talking about the n^2-1 puzzles ? 14:49:31 no 14:50:23 --- join: kc5tja (~kc5tja@66-74-218-202.san.rr.com) joined #forth 14:50:24 --- mode: ChanServ set +o kc5tja 14:50:31 Herkamire> next, the code will go back to the second move and tries the other way. player 1 uses slot 2, plauer 2 uses slot 1. draw reached, game TWO, 6 moves tried 14:50:36 etc etc. 14:50:41 \ 2x2: Moves: 18 Games: 6 14:51:21 mathematically, i'm looking for permutations of 1111222233334444 (for a 4x4), or the formula i just pasted 14:51:49 7 wide by 6 tall seems to be the standard 14:51:54 i know 14:52:07 but robert pasted the number of GAMES that size has 14:52:11 and it had 25 digits 14:52:26 so i wont actually try that board, for now anyways 14:52:27 :) 14:52:51 i've got the whole thing worked out, now i actually need to finalize the code 14:52:58 i have the code to create the games 14:53:05 I don't see why you're mucking around with sql 14:53:06 now altering it to add those queries 14:53:12 seems like a recursion problem to me 14:53:12 because 14:53:22 it is, but i need to work backwards 14:53:31 but i also need to start from the begin 14:53:37 see for the first move 14:53:56 i need to know how many of the next moves of a certain move can result in a win, loss or draw 14:53:58 qFox: there are 6306300 permutations of 1111222233334444 14:54:03 correct 14:54:12 \ 4x4: Moves: 191448940 Games: 63063000 14:54:12 \ Moves: 97552623 Games: 33937264 14:54:12 \ Wins: 10536772 Draws: 23400492 14:54:12 \ Player 1 wins: 5303104 Player 2 wins: 5233668 14:54:27 --- quit: FlamingRain (Connection timed out) 14:54:35 if you take away the "moves after a win/draw" thats the number you end up with 14:54:40 so you can force a win? 14:54:43 yes 14:54:47 that's cool 14:54:49 that has been proven already 14:55:03 --- quit: I440r (Excess Flood) 14:55:04 http://www.ezresult.com/article/Solved_board_games 14:55:14 --- join: I440r (~mark4@216-110-82-59.gen.twtelecom.net) joined #forth 14:55:25 I440r hiya 14:55:35 that's a nice managable number. 14:55:47 its a long term project of mine that i've finally decided to finish 14:55:50 you can fit a byte-array in memory with that many entries 14:56:09 ehm 14:56:18 i can tell you that the results for all games is about 1 gig 14:56:21 1byte per move. 14:56:27 (MOVE, not set of moves) 14:56:34 not storing the moves, storing just who won 14:56:35 1111222233334444 can be considered a game 14:56:42 that serves no purpose 14:56:44 tbw :) 14:56:52 sure it does 14:57:00 that data is useless if you dont know what moves you made prior to winning 14:57:08 unless you're just counting of course 14:57:20 just write a function to map an index in the array to the game move sequence 14:57:45 should be pretty easy 14:58:11 well, i found it much harder then it seemed. thats all i can say 14:58:18 ok 14:58:20 let me think 14:58:27 doing it for tictactoe was much much easier 14:59:17 http://members.home.nl/qfox/4.f 15:00:03 change the 3 to a 2 in the first row of the scanvertical word 15:00:29 changed the parameters but forgot to alter that number at that time 15:00:38 qFox, i think your definitions are too long :) 15:00:47 yeah so noted 15:00:52 not bothered with that tbh :) 15:01:32 also, the counts in the beginning arent accurate for anything heigher then 3, since they were computed using the wrong code 15:01:33 :) 15:01:39 I haven't thought permutations through that well 15:01:53 re TheBlueWizard 15:02:01 * kc5tja is at work, so if I am afk a lot, that's why. 15:02:02 kc5tja hiya 15:02:07 but it seems to me there aught to be a way to convert the permutations to and from unique, consecutive integers 15:02:15 there is 15:02:22 heh i see why you think its so easy 15:02:38 do realize that you actually need to check for a win 15:02:52 kc5tja: I think for X11 over ssh I had to edit /etc/sshd_config 15:02:53 have a board, and scan to see whether a player has 4 in a row, or not 15:03:05 Herkamire: I have that configured. 15:03:17 kc5tja: and the ssh client? 15:03:53 yes 15:04:35 huh 15:04:46 I think on the machines I've used it on, it was enabled by default 15:04:52 so I don't know the stuff you have to do 15:05:01 hold on -- I'm on the phone 15:05:52 Herkamire> consider this example 15:05:53 1 1 1 0 15:05:53 2 2 2 0 15:05:53 1 1 1 0 15:05:53 2 2 2 2 15:05:53 1111222233334000 15:06:04 now the next move permutation wise would be slot 4 15:06:16 however, in the real game this would never occur, since player two won already 15:06:41 yeah, obviously there would be a lot of unused permutations 15:06:42 so the remaining 3 possible moves, are invalid 15:06:58 indeed. this makes the code less simple 15:07:22 although the main body is still relatively simple 15:07:50 you can use a recursive style if you prefer this, but with larger boards you might run out of stackspace 15:07:54 Herkamire: Sorry. Like I said, I'm at work. :) 15:08:06 kc5 is at work ? 15:08:10 where do you work now dood ? 15:08:24 Herkamire: I have the appropriate configurations, and when I do log in, I *do* get localhost:10.0 as the $DISPLAY value. 15:08:31 I440r: kc5tja is working at an ISP 15:09:03 Herkamire: But when I try to run, say, xclock, it refuses to start the program. I forget the precise error message it gives, but I can find *NO* reference to anyone, anywhere, at any time, ever having had this error before. And, of course, zero documentation what-so-ever on it. 15:09:09 qFox: you only need to recurse as deep as the number of moves in a single game, ie 16 on a 4x4 board 15:09:22 awesome! 15:09:24 http://members.home.nl/qfox/4.f is now proper (output is commented out) 15:09:28 thats gotta beat greecyburger 15:09:33 I440r: Same place I have for about two months now; however, like I said before, I punched a hole in my firewall so I can SSH to my box at home. 15:09:48 Herkamire> yes. but for each number you need to remember the board 15:09:57 for each depth i mean 15:10:00 I440r: Huh? You knew I moved out of In-N-Out (not greecyburger; INO is actually a rather nice place to work overall). 15:10:16 the board, the total moves, the current move ("branch" of the tree), etc. 15:10:42 kc5 nope i had NO idea you left there lol 15:10:46 <-- slow 15:10:47 but i have introduction tomorrow (till wednesday) so i'm going to bed now 15:10:58 hf brainstorming if you are :) 15:11:38 I440r: I told you months ago. :) 15:11:44 hi kc5tja 15:11:48 I440r: you can get yourself up to speed, sorta, by checking on kc5tja's website (home page and the blog page) 15:11:58 nah, you could undo moves before returning from a recursion 15:12:25 L) 15:12:27 err 15:12:28 :) 15:12:30 i wonder what i'm doing wrong 15:12:40 --- join: tathi (~josh@pcp02123722pcs.milfrd01.pa.comcast.net) joined #forth 15:12:42 stat()'s st_mode field doesn't match the description in the man page 15:13:00 lol hehe 15:13:05 qFox: in any case, I don't think it would take much memory to store whole coppies of the board for all 42 moves in a 6x7 game 15:13:55 well no, but to solve it i think you'll need gigabytes of free discspace, or a shitload of time with a fast cpu. 15:14:36 anyways i'm off. 15:14:36 nite 15:15:21 --- quit: qFox ("meh ircii's netsplit detection appearantly sux ^^") 15:18:28 --- quit: tathi ("leaving") 15:19:32 grr he left becore i noticed him lol 15:20:44 heh...well, gotta go....all bye! 15:21:05 --- part: TheBlueWizard left #forth 15:30:26 how do you calculate the number of permutations of 112233? 15:31:26 just 6*5*4*3*2 / 8? 15:32:27 I never could figure that stuff out -- it involves factorials, that's all I know. 15:33:12 permutations with unique entries are pretty simple. 15:33:24 for the first digit, you have all 6 to chose from. 15:33:43 so 6 possibilities. for the next digit you have the five you didn't choose to chose from. 15:33:52 so you have 6*5 possibilities. 15:33:58 next place there are four left etc 15:34:24 so for permutations of six unique objects, you have 6*5*4*3*2*1 or simply 6! 15:35:01 I was trying to figure out how you know the number of permutations of something with duplicate entries like 1111222333444 15:35:29 That doesn't make sense. 15:35:46 You have three each of one of four digits, so a total of 12 objects. 15:35:48 That's 12!. 15:36:20 what's 12!? 15:36:21 But if you allow sequences like 111111111111, etc., then the answer would be 4^12. 15:36:46 I just told you what was 12!. 15:37:03 You have 4 kinds of objects (1, 2, 3, 4), and 3 of each kind. That's a total of 12!. 15:37:18 That's a total of 12 rather. 15:37:33 doesn't work that way 15:37:35 For unique permutations, you use the same permutation formula as you did before: 12!. 15:37:38 Why not 15:37:57 it's way too high 15:38:24 No, it follows exactly the same set of laws as your previous 6! example. 15:39:47 ok, you're saying there's 12! orders, 15:39:55 permutations if all were unique 15:39:57 but they aren't 15:40:13 12! is the number of permutations of 12 unique digits 15:40:17 Then your previous example is false for the same reason. 15:40:35 my previous example was just for unique digits 15:40:41 You listed 112233 15:40:49 sorry for the confusion. 15:40:49 There are not 6 unique digits there. 15:41:08 I was starting with an explanation of permutations on unique things 15:41:12 I would think that in that case, there'd be 6!/2 combinations. 15:41:18 Ahh 15:41:53 in the case of 112233, 6! would be way to high, because it would include eg the same looking thing, but with the ones switched 15:43:04 then, I thought, well maybe we could compute the permutations of something with duplicates, if we started with the unique permutations count, and devided it by the number of different orders of things that appear the same 15:43:48 so there's 2 ones, so theres 2! (2) different ways the ones could be rearanged that would look the same. 15:43:59 and same for the twos and threes 15:44:28 --- quit: SDO (Connection reset by peer) 15:44:31 so the total would be: ((6! / 2!) / 2!) / 2! 15:45:03 or 6! / 8 15:45:10 or 90 15:45:35 follow? 15:46:28 so the permutations for 1111222233334444 should be: (((16! / 4!) / 4!) / 4!) / 4! 15:54:28 ahh, and the answer seems to match what qfox and tbw came up with 15:54:39 63063000 16:17:31 Sorry -- was working. 16:17:33 On lunch now though. 16:25:35 --- join: thin (thin@bespin.org) joined #forth 16:29:08 publicly log THIS! 16:30:02 hi thin 16:31:17 THIS! is a C++! construct. We don't talk C++! 'round 'ere. :) 16:31:26 :P 16:31:56 hi kc5tja 16:32:24 hi slava 16:32:25 * kc5tja is --><-- that close to getting FTS/Forth's code to parse words. For the life of me, I can't figure out why it's failing though. I transliterated it to normal GForth, and it works perfectly. This suggests the bug is with the compiler somehow. THis will be nearly impossible to track down. *sigh* 16:35:11 tried gdb? 16:35:30 i heard gdb sucks 16:35:53 thin, yes, for certain things it sucks 16:36:02 it behaves in a counter-intuitive fashion with inline C functions 16:36:06 for assembly i think its fine 16:36:13 hmm 16:36:30 i like being able to interactively execute C functions and look at results, etc. 16:36:30 debuging c is gonna be harder than debugging assembly in any debugger anyways 16:36:36 no its not 16:36:46 well written C/asm/anything is easy to debug 16:37:06 i'm talking about debugging as in using a debugger 16:37:06 C I write on the other hand... *sigh* :) 16:37:15 yes 16:37:20