Author: Lance Perkins
Date: 19:55:44 05/10/05
Go up one level in this thread
May I add something that has not been mentioned before: The "author" claims that the program is 80% in assembly language! 80%!!! For the sake of argument, let's say that Patriot is of comparable size to the source of Toga, since they perform identically. Toga is roughly 13,000 lines of C++ code (not counting the header files). 6,000 of that is for egtb. So, the Patriot code must have 50% of its egtb code in assembly also! Mind you, I don't think anyone has written the egtb code from scratch in C or C++. Now, try that in assembly. How much assembly code are we talking here? So, I let the compiler generate the assembly code for the non-egtb source of Toga. It resulted in 34,000 lines of asm code! Lets say that the 'author' wrote this by hand so its 50% smaller - that would be 17,000 lines of asm code. For all that amount of asm code, it actually plays good chess - bug free, and miraculously identical to an open source program. We have not even considered the egtb source yet. My impression is, the claimed 'author' is either fool enough to undertake the task, or is fool enough to think that he can fool us. Man, if he's that good, Eugene should hire him in the MS VS team to write compiler backends. They could use someone like him in Building 41 in Redmond. --- On May 10, 2005 at 20:05:49, Rolf Tueschen wrote: >to Andy and possible others. Please do NOT think that I want to become expert in >computerchess - I am not! > >But until the case against Patriot 2 and its author finally decided PLEASE give >freedom for all possible arguments or opinions. > >Perhaps in the end I will be proven wrong, but this is my own problem. But I >don't want attend a whitchhunting against a young talent. Until now things could >be explained. But after a final condemnation nobody would be able to argue in >favor of this author. > >If in the end the main "crime" is the marketing of the program, we should not >focus on the programmer alone. Perhaps he didn't know all the details for making >a clone in the end. Perhaps he believed that he took certain models and lost >himself in wishful thinking. > >Please don't give your verdict until the last excuse could be given for all kind >of data... > >To all: Don't worry about me. For me the proof wasn't a proof and until now I'm >not able to understand OTHER evidence that does exist but is only known to real >experts. > >Best to all, Rolf
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.