Rybka-Junior match is off.

[dj]

Dr Wael Deeb wrote:

In the modern world we're living in,there are hundreds of ways to cheat in such a match,

Hundreds This is an important issue. Could you please list (briefly) the main 100 ways to successfully cheat in a match or tournament. I am asking for only a fraction of those you suppose to exist as potentially successful methods.
I am sure we would all welcome your expertise. If even a mere 100 is too onerous then the top 20! Bear in mind what had been agreed about logs in the proposed Rybka-Junior match plus Levy's assessment of this and other matters related to cheating and the match proposals, covered in detail in Levy's pdf file. Or does Levy "make such statements without serious background information" available only to those of your ilk?

Please also explain, if it is "twisted logic" to even contemplate any possible arguments for a match with remote computers, why

1.The recent match in Elista had almost identical conditions to the proposed Rybka-Junior match and the same prize fund?

2. Why is it that almost all past and present matches and tournaments have permitted remote computers?

3. Would you agree that we cannot possibly accept the legitimacy of Rybka's victory in the WCCC this year at Amsterdam? After all Rajlich used a remote computer and "In the modern world we're living in there are hundreds of ways to cheat." With so many ways to cheat surely there must be an element of doubt about Rybka's victory - according to your "logic"?

4. And, of course, you have not answered my previous question about Hydra. Presumably the question was just "twisted logic". Would you ban any combination of hardware and software?

There are, of course, arguments for and against the use of remote computers based on cost, the preferred tuning of a particular computer and issues of cheating, but it cannot possibly be "twisted logic" to suggest that the issues are not black-and white and the conclusions are not self-evident.

[Dr.Wael Deeb]

Dr Wael Deeb wrote:

In the modern world we're living in,there are hundreds of ways to cheat in such a match,

Hundreds This is an important issue. Could you please list (briefly) the main 100 ways to successfully cheat in a match or tournament. I am asking for only a fraction of those you suppose to exist as potentially successful methods.
I am sure we would all welcome your expertise. If even a mere 100 is too onerous then the top 20! Bear in mind what had been agreed about logs in the proposed Rybka-Junior match plus Levy's assessment of this and other matters related to cheating and the match proposals, covered in detail in Levy's pdf file. Or does Levy "make such statements without serious background information" available only to those of your ilk?

Please also explain, if it is "twisted logic" to even contemplate any possible arguments for a match with remote computers, why

1.The recent match in Elista had almost identical conditions to the proposed Rybka-Junior match and the same prize fund?

2. Why is it that almost all past and present matches and tournaments have permitted remote computers?

3. Would you agree that we cannot possibly accept the legitimacy of Rybka's victory in the WCCC this year at Amsterdam? After all Rajlich used a remote computer and "In the modern world we're living in there are hundreds of ways to cheat." With so many ways to cheat surely there must be an element of doubt about Rybka's victory - according to your "logic"?

4. And, of course, you have not answered my previous question about Hydra. Presumably the question was just "twisted logic". Would you ban any combination of hardware and software?

There are, of course, arguments for and against the use of remote computers based on cost, the preferred tuning of a particular computer and issues of cheating, but it cannot possibly be "twisted logic" to suggest that the issues are not black-and white and the conclusions are not self-evident.

You don't need a top 20 methods to do so,read once again Majd's post and you'll figure out what I meant
I have basic knowledge in this field,but I am not a proffesional if this the purpose of your post as I think....

[Kanizsa]

No interest at all. Junior is so weak: Look for CEGT:

Rybka - Deep Junior 10.1 2CPU 33 + 22 = 9 - 2 = 80.3 %

Waste of time...

Jouni

I'm sorry, I don't agree.
All of us know that Amir and Shay are very valuable and creative programmers, but they are not so patient and methodical like SKM to improve slowly Junior during the phasis of the game where Junior is surely weaker.

Looking to the elite, Junior is the weakest program in the endgame phase.
Specific tests show that Junior solves nearly 40 test positions respect to Rybka 2.3.2 that solves 80.

The strenght of Junior is the brilliant conduction of the unbalanced and tactical positions during th middlegame, where the program is second to no one.

The recipe of Amir and Shay seems to me rather clear and reasonabile.
Adding the most powerful hardware they think to improve drammatically the relative weakness during the endgame, cause the brute force may compensate the barely knowledge.

Apart of Zanzibar, Deep Junior is the program which takes the biggest benefit with hardware increasing or infinite analysis setting.

[dj]

Dr Wael Deeb wrote:

In the modern world we're living in,there are hundreds of ways to cheat in such a match,

Hundreds This is an important issue. Could you please list (briefly) the main 100 ways to successfully cheat in a match or tournament. I am asking for only a fraction of those you suppose to exist as potentially successful methods.
I am sure we would all welcome your expertise. If even a mere 100 is too onerous then the top 20! Bear in mind what had been agreed about logs in the proposed Rybka-Junior match plus Levy's assessment of this and other matters related to cheating and the match proposals, covered in detail in Levy's pdf file. Or does Levy "make such statements without serious background information" available only to those of your ilk?

Please also explain, if it is "twisted logic" to even contemplate any possible arguments for a match with remote computers, why

1.The recent match in Elista had almost identical conditions to the proposed Rybka-Junior match and the same prize fund?

2. Why is it that almost all past and present matches and tournaments have permitted remote computers?

3. Would you agree that we cannot possibly accept the legitimacy of Rybka's victory in the WCCC this year at Amsterdam? After all Rajlich used a remote computer and "In the modern world we're living in there are hundreds of ways to cheat." With so many ways to cheat surely there must be an element of doubt about Rybka's victory - according to your "logic"?

4. And, of course, you have not answered my previous question about Hydra. Presumably the question was just "twisted logic". Would you ban any combination of hardware and software?

There are, of course, arguments for and against the use of remote computers based on cost, the preferred tuning of a particular computer and issues of cheating, but it cannot possibly be "twisted logic" to suggest that the issues are not black-and white and the conclusions are not self-evident.

You don't need a top 20 methods to do so,read once again Majd's post and you'll figure out what I meant
I have basic knowledge in this field,but I am not a proffesional if this the purpose of your post as I think....

This is the Ansari post you allude to:

Convekta is the one putting up the money ... so I guess they will have more pull than Vasik on this one. I guess they figure $100,000 is plenty of incentive to try and cheat. Personally I think that the Junior camp would never do such a thing and that they have an impeccable reputation when it comes to honesty. I have no idea why they would disagree to give out log files .... I don't think there is much useful information if any in log files.

Having said that ... Convekta should have the right to ask for equal conditions since they are putting up the money. They have a lot to lose if cheating did occur. They might be thinking that the other side could simply use a certain type of Junior that could use passive output from an existing Rybka engine. This is a project I am working on and it is actually very easy to implement. You run Rybka on a very powerful computer ... then run Junior on another remote computer .... One engine passively checks the moves of the other and when there is a large differential the engine will send that line to engine one. So basically Junior could be playing with Rybka assistance ... and thus play much much stronger. This is very easy to implement and would be impossible to detect as the assisting engine remains passive until a threshold is reached on a certain move. Still that would assume that the other party is motivated to cheat ... and I think that the Amir camp have a very good reputation for being honest.

It seems to suggest one possible method of cheating - not hundreds - and the conditions proposed by the Junior team clearly covered the interesting possibility identified by Ansari. Levy wrote:

The mechanism for this [assuaging fears of cheating] would be the provision of log files and copies of programs, by the programming teams, to the arbiter, after each game. The idea was that, if anything untoward appeared to be happening, the program actually used in the match could be tested.

The underlined sections clearly preemept the sort of cheating suggested by Ansari, which is a fascinating possibility in tournaments where log files and programs are not available for inspection in this way. With such terms and conditions proposed for Rybka-Junior I do not see how one could envisage hundreds of ways of cheating in what would have been 60 minute games.

Of course, there are all sorts of other arguments concerning, for example "fairness" but they are different issues. The cost of having computers on the spot is certainly not insignificant. The cost of hiring hardware for Junior in its New York match against Kasparov in 2003 was between 30,000 and 40,000 dollars. The Rybka-Junior match would have required two lots of hardware and with inflation the cost, as Levy points out, would be more than the prize fund.

[AdminX]

No interest at all. Junior is so weak: Look for CEGT:

Rybka - Deep Junior 10.1 2CPU 33 + 22 = 9 - 2 = 80.3 %

Waste of time...

Jouni

I'm sorry, I don't agree.
All of us know that Amir and Shay are very valuable and creative programmers, but they are not so patient and methodical like SKM to improve slowly Junior during the phasis of the game where Junior is surely weaker.

Looking to the elite, Junior is the weakest program in the endgame phase.
Specific tests show that Junior solves nearly 40 test positions respect to Rybka 2.3.2 that solves 80.

The strenght of Junior is the brilliant conduction of the unbalanced and tactical positions during th middlegame, where the program is second to no one.

The recipe of Amir and Shay seems to me rather clear and reasonabile.
Adding the most powerful hardware they think to improve drammatically the relative weakness during the endgame, cause the brute force may compensate the barely knowledge.

Apart of Zanzibar, Deep Junior is the program which takes the biggest benefit with hardware increasing or infinite analysis setting.

Interesting assessment Franco of Amir and Shay's "match / tournament" strategy for Junior. I never really gave that side of things much thought. However given Junior's dynamic type or style of play if you will, then I must say that that assessment appears to be true.

[Dr.Wael Deeb]

Dr Wael Deeb wrote:

In the modern world we're living in,there are hundreds of ways to cheat in such a match,

Hundreds This is an important issue. Could you please list (briefly) the main 100 ways to successfully cheat in a match or tournament. I am asking for only a fraction of those you suppose to exist as potentially successful methods.
I am sure we would all welcome your expertise. If even a mere 100 is too onerous then the top 20! Bear in mind what had been agreed about logs in the proposed Rybka-Junior match plus Levy's assessment of this and other matters related to cheating and the match proposals, covered in detail in Levy's pdf file. Or does Levy "make such statements without serious background information" available only to those of your ilk?

Please also explain, if it is "twisted logic" to even contemplate any possible arguments for a match with remote computers, why

1.The recent match in Elista had almost identical conditions to the proposed Rybka-Junior match and the same prize fund?

2. Why is it that almost all past and present matches and tournaments have permitted remote computers?

3. Would you agree that we cannot possibly accept the legitimacy of Rybka's victory in the WCCC this year at Amsterdam? After all Rajlich used a remote computer and "In the modern world we're living in there are hundreds of ways to cheat." With so many ways to cheat surely there must be an element of doubt about Rybka's victory - according to your "logic"?

4. And, of course, you have not answered my previous question about Hydra. Presumably the question was just "twisted logic". Would you ban any combination of hardware and software?

There are, of course, arguments for and against the use of remote computers based on cost, the preferred tuning of a particular computer and issues of cheating, but it cannot possibly be "twisted logic" to suggest that the issues are not black-and white and the conclusions are not self-evident.

You don't need a top 20 methods to do so,read once again Majd's post and you'll figure out what I meant
I have basic knowledge in this field,but I am not a proffesional if this the purpose of your post as I think....

This is the Ansari post you allude to:

Convekta is the one putting up the money ... so I guess they will have more pull than Vasik on this one. I guess they figure $100,000 is plenty of incentive to try and cheat. Personally I think that the Junior camp would never do such a thing and that they have an impeccable reputation when it comes to honesty. I have no idea why they would disagree to give out log files .... I don't think there is much useful information if any in log files.

Having said that ... Convekta should have the right to ask for equal conditions since they are putting up the money. They have a lot to lose if cheating did occur. They might be thinking that the other side could simply use a certain type of Junior that could use passive output from an existing Rybka engine. This is a project I am working on and it is actually very easy to implement. You run Rybka on a very powerful computer ... then run Junior on another remote computer .... One engine passively checks the moves of the other and when there is a large differential the engine will send that line to engine one. So basically Junior could be playing with Rybka assistance ... and thus play much much stronger. This is very easy to implement and would be impossible to detect as the assisting engine remains passive until a threshold is reached on a certain move. Still that would assume that the other party is motivated to cheat ... and I think that the Amir camp have a very good reputation for being honest.

It seems to suggest one possible method of cheating - not hundreds - and the conditions proposed by the Junior team clearly covered the interesting possibility identified by Ansari. Levy wrote:

The mechanism for this [assuaging fears of cheating] would be the provision of log files and copies of programs, by the programming teams, to the arbiter, after each game. The idea was that, if anything untoward appeared to be happening, the program actually used in the match could be tested.

The underlined sections clearly preemept the sort of cheating suggested by Ansari, which is a fascinating possibility in tournaments where log files and programs are not available for inspection in this way. With such terms and conditions proposed for Rybka-Junior I do not see how one could envisage hundreds of ways of cheating in what would have been 60 minute games.

Of course, there are all sorts of other arguments concerning, for example "fairness" but they are different issues. The cost of having computers on the spot is certainly not insignificant. The cost of hiring hardware for Junior in its New York match against Kasparov in 2003 was between 30,000 and 40,000 dollars. The Rybka-Junior match would have required two lots of hardware and with inflation the cost, as Levy points out, would be more than the prize fund.

Yes,but note that hthe prices of the hardware or even hiring it had dropped dramatically in the last two years,so this is not a serious issue to be concerned about....
I still insist that there are hundreds of tricks to apply in such a match if you decide to cheat

[Kanizsa]

Looking to the elite, Junior is the weakest program in the endgame phase.
Specific tests show that Junior solves nearly 40 test positions respect to Rybka 2.3.2 that solves 80.

Sorry, my memory failed
http://glareanverlag.wordpress.com/tag/schach/

001. Rybka 2.3.2a 72/100 00:42:26
002. Rybka 2.3.2 71/100 00:45:42
003. Shredder 10 68/100 00:45:17
004. Rybka 2.2 64/100 00:50:48
005. Hiarcs 11.2 64/100 00:54:17
006. Fritz 10 63/100 00:49:47
007. Rybka WinFinder 2.2 61/100 00:48:54
008. Rybka 2.3.1 61/100 00:52:37
009. Deep Frenzee 3.0 61/100 00:54:04
010. Shredder 9.1 59/100 00:56:20
011. Rybka 1.01 B13d 59/100 00:56:31
012. Hiarcs 11.1 59/100 00:58:50
013. The Baron 2.12 58/100 00:59:48
014. SmarThink 1.00 55/100 01:01:32
015. Chess Tiger 2007 54/100 00:58:55
016. Glaurung 2-epsilon/5 54/100 01:03:18
017. Toga II 1.3×4 54/100 01:03:41
018. Spike 1.2 Turin 53/100 00:59:42
019. SlowChessBlitz WV2.1 53/100 01:02:32
020. Glaurung 2-epsilon/2 51/100 01:02:26
021. Delfi 5.1 51/100 01:05:59
022. Ktulu 8 50/100 01:02:55
023. Pharaon 3.5.1 50/100 01:03:32
024. Yace 0.99.87 50/100 01:04:44
025. LoopMP 12.32 50/100 01:06:02
026. Naum 2.0 50/100 01:06:08
027. Fruit 2.3 50/100 01:07:37
028. Fruit 05/11/03 49/100 01:03:33
029. Alaric 707 49/100 01:03:57
030. Loop 10.32f 49/100 01:06:25
031. Hiarcs 10 49/100 01:08:33
032. Ruffian 2.1.0 48/100 01:05:41
033. Toga II 1.2.1a 48/100 01:06:05
034. Fruit 2.2.1 48/100 01:06:29
035. Crafty 20.14 47/100 01:05:26
036. Alaric 704 46/100 01:06:14
037. Colossus 2007a 46/100 01:07:59
038. Glaurung 1.2.1 46/100 01:09:19
039. Deep Sjeng 2.5 46/100 01:09:32
040. Chess Tiger 15.0 46/100 01:10:23
041. Junior 10.1 45/100 01:06:14

------------ooOoo--------
If Junior should be so weak during the endgame (I've no doubt reading this chart), it has to be more strong than Shredder or Fritz during the middlegame, to compensate this gap.

It's known that the endgame phase benefits more than other phasis running on a powerful hardware, but the dynamic unbalanced style of Junior can also benefit from a 16 CPU core in the middlegame, because some of uncorrect lines he likes (for instance the sacrifice ..A:h2+ vs Kasparov) may be cutted off getting deeper and deeper.

No one of us has ever tested Deep Junior 10.1 on a 16 CPU core.

No one knows how Amir and Shay are able to performe efficient high parallelism tecnique in a new version.

It's clear that Convecta fears Junior's potential, both Amir and Shay tecnique and new powerful hardware at the moment unknown.

[dj]

Dr Wael Deeb wrote:

In the modern world we're living in,there are hundreds of ways to cheat in such a match,

Hundreds This is an important issue. Could you please list (briefly) the main 100 ways to successfully cheat in a match or tournament. I am asking for only a fraction of those you suppose to exist as potentially successful methods.
I am sure we would all welcome your expertise. If even a mere 100 is too onerous then the top 20! Bear in mind what had been agreed about logs in the proposed Rybka-Junior match plus Levy's assessment of this and other matters related to cheating and the match proposals, covered in detail in Levy's pdf file. Or does Levy "make such statements without serious background information" available only to those of your ilk?

Please also explain, if it is "twisted logic" to even contemplate any possible arguments for a match with remote computers, why

1.The recent match in Elista had almost identical conditions to the proposed Rybka-Junior match and the same prize fund?

2. Why is it that almost all past and present matches and tournaments have permitted remote computers?

3. Would you agree that we cannot possibly accept the legitimacy of Rybka's victory in the WCCC this year at Amsterdam? After all Rajlich used a remote computer and "In the modern world we're living in there are hundreds of ways to cheat." With so many ways to cheat surely there must be an element of doubt about Rybka's victory - according to your "logic"?

4. And, of course, you have not answered my previous question about Hydra. Presumably the question was just "twisted logic". Would you ban any combination of hardware and software?

There are, of course, arguments for and against the use of remote computers based on cost, the preferred tuning of a particular computer and issues of cheating, but it cannot possibly be "twisted logic" to suggest that the issues are not black-and white and the conclusions are not self-evident.

You don't need a top 20 methods to do so,read once again Majd's post and you'll figure out what I meant
I have basic knowledge in this field,but I am not a proffesional if this the purpose of your post as I think....

This is the Ansari post you allude to:

Convekta is the one putting up the money ... so I guess they will have more pull than Vasik on this one. I guess they figure $100,000 is plenty of incentive to try and cheat. Personally I think that the Junior camp would never do such a thing and that they have an impeccable reputation when it comes to honesty. I have no idea why they would disagree to give out log files .... I don't think there is much useful information if any in log files.

Having said that ... Convekta should have the right to ask for equal conditions since they are putting up the money. They have a lot to lose if cheating did occur. They might be thinking that the other side could simply use a certain type of Junior that could use passive output from an existing Rybka engine. This is a project I am working on and it is actually very easy to implement. You run Rybka on a very powerful computer ... then run Junior on another remote computer .... One engine passively checks the moves of the other and when there is a large differential the engine will send that line to engine one. So basically Junior could be playing with Rybka assistance ... and thus play much much stronger. This is very easy to implement and would be impossible to detect as the assisting engine remains passive until a threshold is reached on a certain move. Still that would assume that the other party is motivated to cheat ... and I think that the Amir camp have a very good reputation for being honest.

It seems to suggest one possible method of cheating - not hundreds - and the conditions proposed by the Junior team clearly covered the interesting possibility identified by Ansari. Levy wrote:

The mechanism for this [assuaging fears of cheating] would be the provision of log files and copies of programs, by the programming teams, to the arbiter, after each game. The idea was that, if anything untoward appeared to be happening, the program actually used in the match could be tested.

The underlined sections clearly preemept the sort of cheating suggested by Ansari, which is a fascinating possibility in tournaments where log files and programs are not available for inspection in this way. With such terms and conditions proposed for Rybka-Junior I do not see how one could envisage hundreds of ways of cheating in what would have been 60 minute games.

Of course, there are all sorts of other arguments concerning, for example "fairness" but they are different issues. The cost of having computers on the spot is certainly not insignificant. The cost of hiring hardware for Junior in its New York match against Kasparov in 2003 was between 30,000 and 40,000 dollars. The Rybka-Junior match would have required two lots of hardware and with inflation the cost, as Levy points out, would be more than the prize fund.

Yes,but note that hthe prices of the hardware or even hiring it had dropped dramatically in the last two years,so this is not a serious issue to be concerned about....
I still insist that there are hundreds of tricks to apply in such a match if you decide to cheat 8-)

I asked only for

1. The top 20 methods of the hundreds of ways to successfully cheat when faced with the terms and conditions proposed for the Rybka-Junior match wth 60 minute games.

2. Why almost all matches and tournaments have allowed, and continue to allow, remote computers - including Rybka's win at Amsterdam.

As for the price of hiring top hardware I was quoting David Levy. He knows better than I do.

[AdminX]

Here is the reply from Sergey Abramov,

http://rybkaforum.net/cgi-bin/rybkaforu ... id=1898#fp

Highlights:

**Initially, we had proposed a “winner-takes-all” approach, i.e. each side should deposit USD 100,000 towards the prize fund and the winner would get the entire prize fund. However, we were aware of Shay Bushinsky’s standpoint, expressed by him in Elista, that such a match “bet” is unacceptable for Junior. The Israeli side was ready to play with guaranteed share in the prize fund only.

**To our mind, remote game is needed for open events to not exclude anybody. For invitational events, this issue disappears. However, Junior’s team insisted on remote game only.

**We wished to avoid in every possible way any suspicions of unfair game and any scandals like that one, which broke out after the ending of the second match between Kasparov and Deep Blue.

[M ANSARI]

Yes, Sergey seems to think that it would be very easy to make the program write logs that do not show cheating. Since they are putting up the money, I guess they have the right to ask for conditions that they are comfortable with. I still think this match will take place somehow.

As for the strength of Junior ... well ofcourse if compared to Rybka, it is very much weaker. But Junior has always seemed to play above its rating in tournaments. Usually the Junior team is very well prepared for matches ... a good example would be its match against Deep Fritz. Deep Fritz probably is the stronger engine, but it was beaten quite soundly by Junior which was much more prepared for the match (better hardware and better opening book).