Uri's Challenge : TwinFish

[pocopito]

Ok, this is the poll I mentioned before:
http://talkchess.com/forum/viewtopic.ph ... viewresult

[Milos]

The only changes made to reach a "<55%" similarity are a complete asymetric PST (based on Adam Hair values).

If you want to see the changes just search for "Robber" in the sources files.

This is well known thing from long ago, that similarity test actually measures PST matching. All other eval terms are completely irrelevant.
It's totally unscientific thing, made to look like some science.

[Milos]

And so we are witnessing the death of similarity tester. Now that the cat is out of the bag I can confirm Ben's findings. During the PST-thread in the programmers forum I did some experiments with the several posted PST's and Piece Values and indeed they dreadfully bring down the similarity percentage without too much elo loss (20-30).

So folks be aware, cloners will find out anyway.

Still, no false positives with Sim, only false negatives.

That's just ridiculous, how would you prove false positive?
Even if you had source code, could you prove functional equivalence?

[tpetzke]

It still puzzles me why people develop private engines (so you can't even run the similarity test?) or closed source engines when they are hundreds of elo below the top engines. Why do they fear to show us their code?

Why do you think it's fear. I just like the concept of property. I have no problem in sharing my ideas or tests or concepts that I have implemented. I just don't share my code. Someone really interested in chess programming doesn't need source code beyond the stuff already contained in the wiki to keep going.

If in the crafty sources all code was removed and only the comments remained it would not be any less useful to me.

I find the concept of polluting my engine with code from others rather disturbing. It provides only value to me as it is clean, so I know every win is my win and every loss is my loss. If someone finds that suspicious I can live with that.

[Steve Maughan]

(...) I just like the concept of property (...)

+1

Indeed!

Steve

[Steve Maughan]

(...)The similarity is less than 55% and the elo fall is about 70-80 only.(...)

I fear a new era of clones!

And yes, I believe there are some people who would be happy to say they "created" a chess engine which is "almost as good" as the World's best.

Steve

[Steve Maughan]

OK - how about the reverse challenge?

Can you take a weak open source engine (say Maverick) and change the piece-square-tables such that it now *FAILS* a similarity test (>60%).

Steve

[Andres Valverde]

(...) I just like the concept of property (...)

+1

Indeed!

Steve

+2

[velmarin]

I've never seen positions used in SIM.
Although it seems they can be changed.

They will guess middlegame.
If you change positions near the opening, what would happen?
or change them at the end positions.

[michiguel]

It still puzzles me why people develop private engines (so you can't even run the similarity test?) or closed source engines when they are hundreds of elo below the top engines. Why do they fear to show us their code?

Why do you think it's fear. I just like the concept of property. I have no problem in sharing my ideas or tests or concepts that I have implemented. I just don't share my code. Someone really interested in chess programming doesn't need source code beyond the stuff already contained in the wiki to keep going.

If in the crafty sources all code was removed and only the comments remained it would not be any less useful to me.

I find the concept of polluting my engine with code from others rather disturbing. It provides only value to me as it is clean, so I know every win is my win and every loss is my loss. If someone finds that suspicious I can live with that.

The question should be the opposite, why should be open the code? It takes time to do that. If there is a good reason, I open it. Otherwise, I do the default, which is, I do not[1]. For the tablebases, I was convinced there was a good reason to open it and I released under the MIT license. Same with ordo, I saw a purpose and I opened it. I thought about opening Gaviota for years, but I see little purpose in open it and frankly, I see more problems in the horizon than anything else. If anybody think that cloning a non-top engine is unlikely, that person would be very wrong. It happened many times. When there was something that I thought it was worth sharing, I verbalized it here in forum.

In addition, why do we assume that nobody has seen the code of closed engines except the author? people share things privately. Dann Corbit has compiled some of gaviota versions for years, made suggestions etc. etc.. The concept that people do not GPL/MIT their code because they are afraid of something is uni-dimensional. Some things are suitable for OS, some are not. I do not believe in rigid dogmas.

Miguel
[1] In fact, many times I wondered why I even released binaries, but there are good friends who enjoy it, and it is a good idea for programmers of non-top engines to have more partners.