Longer Article

[Albert Silver]

The point isn't that it was done in innocence, the point is that Cozzie implied that the purpose was to hide the fact it was a clone ("only done by clones"). To be honest, I never heard of a clone lying about the ply-depth, and I cannot begin to imagine how the ply-depth might be considered revealing about a program's origins.

If we modify Rybka 1.0 and add 3 to its ply-depth, what does that show you about its origins? The answer is: Nothing.

It doesn't prove anything by itself. The first thing you want to do when verifying whether a program A is derived from program B is compare the output of both programs using some test positions. If A's output is obfuscated (depth/nps/score) it is more difficult to link A with B by only observing it. Simple.

You never heard about clones obfuscating its output? You don't need to go far. It seems earlier versions of Houdini (Not sure about 1.5) obfuscates the score multiplying it by a factor of ~0.5.

Obviously you are not going to fool an expert by doing this, but it "helps" a bit.

Some links about Houdini obfuscating outputs:
http://talkchess.com/forum/viewtopic.ph ... t&start=60
http://www.talkchess.com/forum/viewtopi ... =&start=40

Regards,

Like I said, I don't dispute the ply-depth obfuscation by Rybka, just the purpose. I honestly don't think it was done to cover up its origins, I think it was to cover the fact that it was simply outsearching other engines by a huge margin, giving the impression its crushing results were the sole consequence of some mind-boggling evaluation function.

[Uri Blass]

If the target is to hide the orign of the program then it is easy to find smarter ways then multiplying the evaluation by a constant.

It is better to use another monotonic function so it is less easy for people to notice the relation easily by watching the output(it is also possible to add random number to the evaluation that the computer shows but in that case people are going to complain that the program is useless for analysis).

My guess is that the reason for multiplying rybka's evaluation by some factor that is close to 0.5 is that some pruning is more productive when you do it and not in order to hide houdini's origin.

[JuLieN]

My guess is that the reason for multiplying rybka's evaluation by some factor that is close to 0.5 is that some pruning is more productive when you do it and not in order to hide houdini's origin.

Uri, I am a jurist, not a mathematician, but I think they taught me at school that a "multiplication by 0.5" is actually called a "division by 2"

[Tom Barrister]

Well , my question ( in this thread) was about the big jump in ELO points of Hiarcs , the commercial army & the top free engines- after Rybka publication.

I wasn't aware that you were asking me that question. But to answer, Hiarcs has gained 300 ELO in six years. I would venture that computer programs that were around then and are around now have gained the same or more, on average, in six years, so Hiarcs gain is probably due to the advances in hardware technology and freely-available software improvements during the same period.

And , OK , I think Mr.Rajlich aren't too interested in his no. 4 version. The cluster version monopolised him.

Mr. Rajlich should be paying more attention to his existing customers than he does.

[Tom Barrister]

Disassembly of commercial programs is commonly disallowed by license agreements.

I also made this point elsewhere. I'll sum it up again; this applies to the United States and probably most other countries: Reverse engineering (or whatever other term you use) is okay, unless the license agreement restricts it. If the license restricts it, then by reverse engineering it, the contract with the software's copyright owner is null and void. This makes the person who is reverse engineering guilty of copyright infringement, since he/she is no longer a legal user of the software. This later point is important, because it allows the case to be moved from the state court to the Federal court system and may allow criminal charges to be filed.

[Sylwy]

Hi,


No more interviews or updates on the Arena page till they learn to cite ask for permission and cite the entire interview.
This is why you ..........................................................
Michael
http://www.playwitharena

Miky

Do you have a fix ?
I'm not interested at all with your interviews on Arena web page.
Even I deleted ( me and my friends ) Arena GUI -from all my computers- after your insults (the bibliography is here: http://talkchess.com/forum/viewtopic.ph ... 27&t=38190 ).
Also I have nothing to do with the publication -without quoting- of your interviews on various chess forums. Ask for the right person ( every post, in every forum , has a left case with the name/nick of the poster) , please !
So , where's the problem ?

S

[Michael Diosi]

Hi,

Maybe Martin will make something, I will do as little as possible. Maybe nothing at all. The domain is payed untill July so let's see what will happen after that.

You and the people there expect respect but show none. Which such a behaviour you will be waiting for the plumber the next 20 years. And what name are you talking about, they don't have the spine to use their full real name. It is upto you if you want to talk to Indianajones & Co.

The hits and downloads from that part of the world talk a different language. Also there are some people with spine who say the same I do.

Michael
http://www.playwitharena.com

[Sylwy]

Hi,

Maybe Martin will make something, I will do as little as possible. Maybe nothing at all. The domain is payed untill July so let's see what will happen after that.

You and the people there expect respect but show none. Which such a behaviour you will be waiting for the plumber the next 20 years. And what name are you talking about, they don't have the spine to use their full real name. It is upto you if you want to talk to Indianajones & Co.

The hits and downloads from that part of the world talk a different language. Also there are some people with spine who say the same I do.

Michael
http://www.playwitharena.com

Blah-blah-blah !

A single link with my real name/nick and one of your interviews please.
Otherwise someone is:



No digressions and evasive posts. Concrete : me and your interviews links.
Your insults are here (the links):

http://talkchess.com/forum/viewtopic.ph ... 27&t=38190
http://talkchess.com/forum/viewtopic.ph ... 09&t=38190

I point only two !
Everybody here can judge if someone is a honest person ! Or not !

S

[Michael Diosi]

Hi,

Links to the forum there are not allowed here. I am sure your able to find them by yourself (maybe I overestimate your abbilities, sorry if that is the case). The forum does have a "search" option ? I have sent the links to one of the moderators here and he has seen them.

Yes, everybody can judge that they are not able to cite something or ask for permission.

Thank you for bringing this up again.

Michael
http://www.playwitharena.com

[Sylwy]

Yes, everybody can judge that they are not able to cite something or ask for permission.

http://www.playwitharena.com

I ask about "me" not "they". Digressive again ?

What to talk (chess) here a bit about "they". Maybe you have known persons "there" ? Maybe. For example somebody "there" loves a lot ChessBase products. And...loves Romanians also.At least one. Like you ( the link is here:
http://talkchess.com/forum/viewtopic.ph ... 27&t=38190 ).

What about ?

S