Author: Uri Blass
Date: 15:06:41 01/23/06
Go up one level in this thread
On January 23, 2006 at 17:27:39, Joseph Tadeusz wrote: >On January 23, 2006 at 17:17:17, Dann Corbit wrote: > >>On January 23, 2006 at 16:59:33, Joseph Tadeusz wrote: >> >>>On January 23, 2006 at 15:39:04, Dann Corbit wrote: >>> >>>>On January 21, 2006 at 20:13:57, Uri Blass wrote: >>>> >>>>>On January 21, 2006 at 20:09:43, Uri Blass wrote: >>>>> >>>>>>On January 21, 2006 at 17:53:53, David H. McClain wrote: >>>>>> >>>>>>>On January 21, 2006 at 12:45:04, Rolf Tueschen wrote: >>>>>>> >>>>>>>>Please read here and join the debate: >>>>>>>> >>>>>>>>http://www.talkchess.com/forums/2/message.html?203547 >>>>>>> >>>>>>>Rolf, >>>>>>> >>>>>>>It may be everything stated in that post but I am not naive enough to think >>>>>>>Rybka has not been disassembled and reverse engineered by as many professional >>>>>>>and many amatuer authors you can think of, Russian or otherwise. >>>>>>> >>>>>>>There is money at stake. DHM >>>>>> >>>>>>I do not think that there is much money in chess. >>>>>> >>>>>>Everybody that I talk with him tell me that people who can earn money from chess >>>>>>programs can earn a lot more from other things. >>>>>> >>>>>>I also doubt if disassembling rybka is the the best way to get a strong chess >>>>>>program and it is possible that people who are so smart to be able to do it and >>>>>>understand the assembler code of rybka may be also smart enough to generate >>>>>>something better in less time. >>>>>> >>>>>>Uri >>>>> >>>>>I can add that in the past genius dominated for years by a big margin. >>>>> >>>>>people also could disassemble genius at that time and in that time it was >>>>>possible to earn more money from chess programming. >>>>> >>>>>The fact that for years no program came close to genius suggest that >>>>>disassembling and reverse engineering is not so easy. >>>> >>>>I think that for a large system (e.g. 500K exe or bigger) it will be harder to >>>>disassemble and recreate than to write from scratch. >>>> >>>>Disassembled binary instructions will form a huge volume of information. >>>> >>>>A few lines of C can expand into a large volume of assembly. >>> >>> >>>There are tools that translate executables directly into C code. >> >>DCC and Boomerang are the two that I am aware of. They spit out unintelligible >>goo. > >I don't agree. I think the windows version of DEC creates high quality C code. >All that remains are the following steps: > > - identify important functions > - identify important variables In that case I guess that reverse engineering of fruit is easier because you can reverse engineering both fruit2.1 and fruit2.2 and for fruit2.1 you already has the name of important functions and important variables. In that case I am surprised that almost all the clones that were discovered were clones of free programs with source and not clones of commercial programs because clonners can also reverse engineer commercial programs and change the code and claim to have a new program. This is the main reason that I would like to see a situation when reverse engineering is simply practically impossible(doing it illegal is not going to stop cloners who do not care about the law). Another good solution may be if there is a program that take 2 exe files and get correct conclusions if one of them is based on code of the other program. The program may also return do not know but I want it always to return correct results and to be effective at least in finding that toga is based on fruit and that Crafty-20 is based on crafty-17 without detecting false results when it compare between Crafty and fruit or other cases. Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.