Dann Corbit wrote:Now Vas can take a look and know for sure what similarities there are (if he is even interested)
He already took a look into Strelka's source (before it was released) and claimed that it was a clone of Rybka BETA.
This is simply not true.
This is quote from Osipov in russian:
Странный он - этот Васик.
Три раза я предлагал ему ознакомиться с исходными текстами Стрелки - в июле, сентябре и октябре. В июле ему это было неинтересно, интересовало только - насколько трудно было изучать алгоритм Рыбки дизассемблером, и насколько труднее это в отношении новых версий Рыбки. Я ему подробно обо всем написал.
В сентябре он был очень занят подготовкой к матчу с Заппой, и ему было не до Стрелки. После моего октябрьского письма он думал о чем-то недели три, и ответил как-то совсем туманно.
И вот теперь это его последнее заявление... Непонятно - исходники Стрелки читать не хотим, но мнение о ней имеем твердое.
here it is in English(machine translation, but quite readable):
Strange person is he - this Vasik.
Three times I proposed him to view the source code of Strelka - in July, September and October. In July it it was uninteresting for him, he was interested only how difficult it was to study algorithm of the Rybka by disassembling it, and if it is more difficult with the new versions of the Rybka. I gave him detailed answer. In September he was very busy with preparation for a match with Zappa, and he had no time for Strelka. After my letter from October he was silent for three weeks, and has answered somehow quite mistily.
And now here is his last statement... Incomprehensible - he does'nt want to read the source code of Strelka, but his opinion about it is quite hard.
I remember that
Vasik claimed in the rybka forum that he has the strelka code so I guess that he got the code.
John Conway wrote:Vasik claims Strelka is a Rybka 1.0 clone and he doesn't care about that. Now that Strelka is open source, does this mean that Rybka code is exposed?
Dann Corbit wrote:
I guess that the similarities in code would barely be recognizable.
Vas stated catagorically that Strelka is "obviously" a Rybka 1.0 clone, so the similarities in the code should be easily recognizable to him, and he could clarify the similarities for others to see.
That is very good. I think it removes many questions as far as Fruit is concerned, since it is now open source (although the license is not clear).
Now Vas can take a look and know for sure what similarities there are (if he is even interested) and I also invite other chess authors to compare with Fruit and I think that they will also come to the same conclusions that I came to (clearly inspired by Fruit, but not simply a clone).
Uri has found lots of interesting ways to shrink the code of the engine. He may wish to add a fork now, if that seems pleasing to him.
I see that the code is already smaller and simpler and what I found is probably not relevant because strelka already got rid of the ugly random numbers.
Uri
You claimed before that the only tables you could find were these random tables. Yet the sourcecode contains tables for lots of different piece configurations.
That is very good. I think it removes many questions as far as Fruit is concerned, since it is now open source (although the license is not clear).
Now Vas can take a look and know for sure what similarities there are (if he is even interested) and I also invite other chess authors to compare with Fruit and I think that they will also come to the same conclusions that I came to (clearly inspired by Fruit, but not simply a clone).
Uri has found lots of interesting ways to shrink the code of the engine. He may wish to add a fork now, if that seems pleasing to him.
I see that the code is already smaller and simpler and what I found is probably not relevant because strelka already got rid of the ugly random numbers.
Uri
You claimed before that the only tables you could find were these random tables. Yet the sourcecode contains tables for lots of different piece configurations.
You didn't recognize these ?
Tony
Maybe there was a misunderstanding.
I do not remember claiming that there were not tables for a lot of different piece configurations.
I also do not remember claiming that the only tables were tables of random numbers.
some big tables in the old code that was bigger were xored with random number so they were seen as tables of random numbers and it took me time to figure them out and write special functions to calculate them
This job is now done independetly in a different way in the function init_bitboards() in board.c.
I did not check it but it seems to me that my code for the same purpose was longer.
// âåðèôèêàöèÿ ïóñòîãî õîäà - òîæå èç îäíîé èç Áåëîê, ÷òî äàåò - íåèçâåñòíî.
// Die Verifizierung des leeren Laufs(Gangs) - auch aus einem des Eiweisses, was gibt - ist es nicht bekannt
// Verification of an empty course - too from one of Fiber that gives - it is not known
// âåðèôèêàöèÿ ïóñòîãî õîäà - òîæå èç îäíîé èç Áåëîê, ÷òî äàåò - íåèçâåñòíî.
// Die Verifizierung des leeren Laufs(Gangs) - auch aus einem des Eiweisses, was gibt - ist es nicht bekannt
// Verification of an empty course - too from one of Fiber that gives - it is not known
Babel Fish says ...
//the verification of empty motion - also of one of the squirrels, what gives - unknown.
// âåðèôèêàöèÿ ïóñòîãî õîäà - òîæå èç îäíîé èç Áåëîê, ÷òî äàåò - íåèçâåñòíî.
// Die Verifizierung des leeren Laufs(Gangs) - auch aus einem des Eiweisses, was gibt - ist es nicht bekannt
// Verification of an empty course - too from one of Fiber that gives - it is not known
Babel Fish says ...
//the verification of empty motion - also of one of the squirrels, what gives - unknown.
Regards
Dave
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
