Rybka Coding Posts

Discussion of anything and everything relating to chess playing software and machines.

Moderator: Ras

User avatar
Graham Banks
Posts: 44654
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Rybka Coding Posts

Post by Graham Banks »

tiger wrote:
Graham Banks wrote:
tiger wrote: Tell me, Graham, do you remember a few cases where an engine has been banned from further tesing by the CCRL? Which ones and why?



// Christophe
We stopped testing Strelka. That's all.


I assume you did so because Strelka was convicted of something by some legal official entity?



// Christophe
No - we made the decision ourselves to stop testing it.
gbanksnz at gmail.com
rebel777

Re: Rybka Coding Posts

Post by rebel777 »

Hi Pedro,
pedrox wrote: I can look at CW and Ed as two special programmers to which I did not get either the shoe, but comments on disassembly programs by Chris and acceptance by Ed makes no sense, is completely ridiculous.
If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
pedrox wrote: If the disassembly allowed Yury Osipov build Strelka from assembler code of Rybka and for Vas Strelka is a clone, then also make it possible to compare the disassembly of Rybka and Fruit.
Of course, nobody doubts that. Point is that good data isn't presented yet. Still the accusations continue.
pedrox wrote: If we have to see the disassembly as something artistic, then Yury Osipov is the greatest artist of all.
He for sure is :lol: :lol: Fact remains reengineering is creative process which includes mistakes. Others should be able to judge. So far there is nothing to judge.

Ed
Uri Blass
Posts: 10899
Joined: Thu Mar 09, 2006 12:37 am
Location: Tel-Aviv Israel

Re: Rybka Coding Posts

Post by Uri Blass »

tiger wrote:
Graham Banks wrote:
tiger wrote: Tell me, Graham, do you remember a few cases where an engine has been banned from further tesing by the CCRL? Which ones and why?



// Christophe
We stopped testing Strelka. That's all.


I assume you did so because Strelka was convicted of something by some legal official entity?



// Christophe
I think that in the case of strelka the public opinion was that it is a clone.
I posted evidence for identical behaviour to rybka.

There are clearly more evidence for similiarity between rybka and strelka relative to the case of rybka and fruit.

In the case of rybka and strelka1.8 you do not need the source code and you can simply see identical behaviour with rybka based on analysis
including sharing some bugs that fruit does not have.

Here is an example for common bug that only rybka and strelka share
together with identical behaviour.
You can see significant change in the score from depth 1 to depth 7 with no tactical reason

scores of strelka1.8 at depth 5-10 are identical to score of rybka
at depths 3-8

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNB1KBNR w KQkq - 0 1

Analysis by Strelka 1.8 UCI:

1.Ng1-f3
-+ (-6.41) Depth: 1 00:00:00
1.Nb1-c3 Nb8-c6
-+ (-6.97) Depth: 2 00:00:00
1.Nb1-c3 Nb8-c6 2.Ng1-f3
-+ (-6.41) Depth: 3 00:00:00
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6
-+ (-6.97) Depth: 4 00:00:00
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6 3.d2-d3
-+ (-6.73) Depth: 5 00:00:00
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6 3.d2-d3 d7-d5
-+ (-6.94) Depth: 6 00:00:00 3kN
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6 3.d2-d3 d7-d5 4.a2-a4
-+ (-9.00) Depth: 7 00:00:00 18kN
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6 3.d2-d3 d7-d5 4.a2-a4 Nc6-b4
-+ (-8.94) Depth: 8 00:00:00 569kN
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6 3.d2-d3 d7-d5 4.a2-a4 Nc6-b4 5.Ke1-d1
-+ (-9.00) Depth: 9 00:00:02 807kN
1.Nb1-c3 e7-e6 2.Ng1-f3 Ng8-e7 3.d2-d3 Ne7-d5 4.Nc3-b1 Bf8-b4+ 5.Ke1-d1 Nb8-c6 6.Nb1-a3
-+ (-9.00) Depth: 10 00:00:05 3534kN

(so k, 01.09.2008)

[d]rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNB1KBNR w KQkq - 0 1

Analysis by Rybka 1.0 Beta 32-bit:

1.Nb1-c3
-+ (-6.73) Depth: 3 00:00:00
1.Nb1-c3
-+ (-6.94) Depth: 4 00:00:00
1.Nb1-c3 Nb8-c6
-+ (-9.00) Depth: 5 00:00:00
1.Nb1-c3 Nb8-c6 2.Ng1-f3
-+ (-8.94) Depth: 6 00:00:00 163kN
1.Nb1-c3 Nb8-c6 2.Ng1-f3 Ng8-f6
-+ (-9.00) Depth: 7 00:00:00 227kN
1.Nb1-c3 d7-d5 2.e2-e3 Nb8-c6 3.Ng1-f3
-+ (-9.00) Depth: 8 00:00:02 699kN

(so k, 01.09.2008)




Uri
User avatar
Graham Banks
Posts: 44654
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Rybka Coding Posts

Post by Graham Banks »

rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
gbanksnz at gmail.com
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Rybka Coding Posts

Post by Terry McCracken »

Graham Banks wrote:
rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
No, it's not well stated. In fact it totally misses the mark altogether.

If we wait to do this behind closed doors then we'll be long dead. That's not realistic.

Even if behind closed doors and 10 years from now a verdict is reached so to speak it's far too late.

Also, this would eventually spill over and appear more evil in many people's eyes.

This is a win-win lose-lose proposition.
User avatar
Graham Banks
Posts: 44654
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Rybka Coding Posts

Post by Graham Banks »

Terry McCracken wrote:
Graham Banks wrote:
rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
No, it's not well stated. In fact it totally misses the mark altogether.

If we wait to do this behind closed doors then we'll be long dead. That's not realistic.

Even if behind closed doors and 10 years from now a verdict is reached so to speak it's far too late.

Also, this would eventually spill over and appear more evil in many people's eyes.

This is a win-win lose-lose proposition.
So you prefer vigilante justice to going through proper channels?
gbanksnz at gmail.com
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Rybka Coding Posts

Post by Terry McCracken »

Graham Banks wrote:
Terry McCracken wrote:
Graham Banks wrote:
rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
No, it's not well stated. In fact it totally misses the mark altogether.

If we wait to do this behind closed doors then we'll be long dead. That's not realistic.

Even if behind closed doors and 10 years from now a verdict is reached so to speak it's far too late.

Also, this would eventually spill over and appear more evil in many people's eyes.

This is a win-win lose-lose proposition.
So you prefer vigilante justice to going through proper channels?

No Graham, and that isn't happening. George apologised and bowed out.

I suggest for the sake of discussion you do the same.
User avatar
Graham Banks
Posts: 44654
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Rybka Coding Posts

Post by Graham Banks »

Terry McCracken wrote:
Graham Banks wrote:
Terry McCracken wrote:
Graham Banks wrote:
rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
No, it's not well stated. In fact it totally misses the mark altogether.

If we wait to do this behind closed doors then we'll be long dead. That's not realistic.

Even if behind closed doors and 10 years from now a verdict is reached so to speak it's far too late.

Also, this would eventually spill over and appear more evil in many people's eyes.

This is a win-win lose-lose proposition.
So you prefer vigilante justice to going through proper channels?

No Graham, and that isn't happening. George apologised and bowed out.

I suggest for the sake of discussion you do the same.
I have nothing to apologise for Terry.
I repeat that if Christophe, Zach and company think they have a case, they should take it to the FSF.
That they seem reluctant to do so speaks volumes.
gbanksnz at gmail.com
Terry McCracken
Posts: 16465
Joined: Wed Aug 01, 2007 4:16 am
Location: Canada

Re: Rybka Coding Posts

Post by Terry McCracken »

Graham Banks wrote:
Terry McCracken wrote:
Graham Banks wrote:
Terry McCracken wrote:
Graham Banks wrote:
rebel777 wrote: If you go through my postings you will notice I have argued neither pro or against Vas. I have said the procedure is wrong. It's wrong to spread accusations without full proof. I also argued this discussion should be done behind the scenes (by email) and not in public. Now the whole thing smells as a smear campaign whether intended or not. Fact is harm is done to Vas' good name. No matter how the verdict will be it's the wrong procedure.
Ed
Exactly. Well stated Ed.
No, it's not well stated. In fact it totally misses the mark altogether.

If we wait to do this behind closed doors then we'll be long dead. That's not realistic.

Even if behind closed doors and 10 years from now a verdict is reached so to speak it's far too late.

Also, this would eventually spill over and appear more evil in many people's eyes.

This is a win-win lose-lose proposition.
So you prefer vigilante justice to going through proper channels?

No Graham, and that isn't happening. George apologised and bowed out.

I suggest for the sake of discussion you do the same.
I have nothing to apologise for Terry.
I repeat that if Christophe, Zach and company think they have a case, they should take it to the FSF.
That they seem reluctant to do so speaks volumes.
They are gathering data. That's all. You really don't understand and am through trying to explain this to you.

It's pointless as you don't see what the real picture is and you and others don't wasnt to see it.

I have to leave. I suggest you think before you reply to me.
User avatar
Graham Banks
Posts: 44654
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Rybka Coding Posts

Post by Graham Banks »

Terry McCracken wrote:
Graham Banks wrote:I have nothing to apologise for Terry.
I repeat that if Christophe, Zach and company think they have a case, they should take it to the FSF.
That they seem reluctant to do so speaks volumes.
They are gathering data. That's all. You really don't understand and am through trying to explain this to you.

It's pointless as you don't see what the real picture is and you and others don't wasnt to see it.

I have to leave. I suggest you think before you reply to me.
Good. They should gather their data privately and then present their case.
Posting here in the meantime achieves little except draw fire.
gbanksnz at gmail.com