From cube-lovers-errors@mc.lcs.mit.edu Tue Feb 15 18:33:33 2000 Return-Path: Received: from sun28.aic.nrl.navy.mil (sun28.aic.nrl.navy.mil [132.250.84.38]) by mc.lcs.mit.edu (8.9.1a/8.9.1-mod) with SMTP id SAA05807 for ; Tue, 15 Feb 2000 18:33:33 -0500 (EST) Precedence: bulk Errors-To: cube-lovers-errors@mc.lcs.mit.edu Date: Thu, 27 Jan 2000 21:38:47 -0500 (EST) From: der Mouse Message-Id: <200001280238.VAA02190@Twig.Rodents.Montreal.QC.CA> To: Ghan Cc: Cube-Lovers@ai.mit.edu Subject: Re: Stuff... > I don't really know whether most people on this mailing list are > regulars or just a bunch of strangers who don't know each other, My sense of it is that - like most lists - cube-lovers consists of a few relatively prolific posters and a whole lotta lurkers. But only the listowner really knows.... > but I'll introduce myself anyway. [...] Welcome! Glad to see you here. > [L]ast July, I got one of my friends to teach me how to solve the > Rubik's Cube, which I have ever since regretted, since I really would > have liked to solve it myself... Heh. Yeah, that's a danger. I managed most of it myself, though it took a while; the one piece someone had to show me is a double edge flip. > Anyway, I have quite a few unrelated questions I need help with... > What type of glue could I use to secure all the stickers more > permanently? If you don't mind something really permanent, you can always take the stickers off and paint the facicles. I've done this on a 3-Cube and have a 5-Cube sitting disassembled waiting for the same treatment. > Finally, I decided today to try my hand at some programming. > [...exhaustive search for the 2-Cube...] > I thought I was pretty good until I discovered that my program took > 18 minutes to check all the 9 move algorithms! The 2-Cube *has* been completely solved; back in 1992 I wrote a program that did so, and I'm sure dozens of others have done likewise, before and since. Here are position counts: For the quarter-turn metric: 1 at distance 0 [Wed May 13 23:18:51 1992] 6 at distance 1 [Wed May 13 23:18:52 1992] 27 at distance 2 [Wed May 13 23:18:53 1992] 120 at distance 3 [Wed May 13 23:18:54 1992] 534 at distance 4 [Wed May 13 23:18:55 1992] 2256 at distance 5 [Wed May 13 23:18:56 1992] 8969 at distance 6 [Wed May 13 23:18:58 1992] 33058 at distance 7 [Wed May 13 23:19:06 1992] 114149 at distance 8 [Wed May 13 23:19:29 1992] 360508 at distance 9 [Wed May 13 23:20:53 1992] 930588 at distance 10 [Wed May 13 23:24:52 1992] 1350852 at distance 11 [Wed May 13 23:35:53 1992] 782536 at distance 12 [Wed May 13 23:53:44 1992] 90280 at distance 13 [Thu May 14 00:03:19 1992] 276 at distance 14 [Thu May 14 00:04:25 1992] For the half-turn metric: 1 at distance 0 [Thu May 14 00:12:37 1992] 9 at distance 1 [Thu May 14 00:12:39 1992] 54 at distance 2 [Thu May 14 00:12:41 1992] 321 at distance 3 [Thu May 14 00:12:42 1992] 1847 at distance 4 [Thu May 14 00:12:44 1992] 9992 at distance 5 [Thu May 14 00:12:56 1992] 50136 at distance 6 [Thu May 14 00:13:08 1992] 227536 at distance 7 [Thu May 14 00:14:26 1992] 870072 at distance 8 [Thu May 14 00:20:43 1992] 1887748 at distance 9 [Thu May 14 00:47:06 1992] 623800 at distance 10 [Thu May 14 01:46:46 1992] 2644 at distance 11 [Thu May 14 02:07:51 1992] I don't recall why the half-turn run was so much slower; perhaps something else (backups?) started running at midnight.... > [T]here are computer programs which can solve the 3x3x3 in seconds > (or so I've read, There are - but not if you also demand that the result be certain to be an optimal solution. Heck, a program could be written that just cranks through a recipe such as you can find in any of dozens of books; such a program could run in less than a second on all but the tiniest of computers. The interest lies in finding *good* - or otherwise interesting - solutions. > So I'm wondering how to write a fast solver. Is my program slow > because of my bad programming skills, or is my method just plain > slow? The latter, certainly. The former, well, perhaps; I haven't seen the code, so it's hard to say. :-) Certainly your choice of language will work against producing a high-speed program. der Mouse mouse@rodents.montreal.qc.ca 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B