Hi Damien Right, interesting only for the theory. But also a lot of fun to play with.
Did you see my example of a TM for binary addition of two arbitrary large numbers? It has 13 states (of which only 9 are essential) and 51 rules (of which only 27 are essential) The test includes 50 tests with binary encodings of numbers (random 30000000). All tests together take about 5 seconds on my simple PC, (conversion between exact-nonnegative-integers and their binary encodings included) https://github.com/joskoot/turing. It took me two days to find out why the UTM presented by Hopcroft and Ullman does not work (in their 1969 edition of "Formal Languages and their Relation to Automata") May be these errata have been corrected in more recent editions, but I have no free access to them to check this. May be someone can tell me? Thanks for your reply, Jos -----Original Message----- From: racket-users@googlegroups.com [mailto:racket-users@googlegroups.com] On Behalf Of Damien MATTEI Sent: lunes, 12 de junio de 2017 14:36 To: racket-users@googlegroups.com Subject: Re: [racket-users] Turing machines snip Interesting but from a theoretical point of view, when i was student i programmed a Turing machine emulator in PASCAL (still somewhere on the net) , and as example i started to code in Turing states a 8bit addition number, i stopped near the 4th bits, there was so many states... it is interesting for students but for the theory only. Damien -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
