Alfil 15.7

[Dann Corbit]

Version with source and projects, along with correct icon and CPU type of Any(switches as appropriate) and X64 (hardwired to 64 bit)
Uses dotnet 4.0

https://www.dropbox.com/s/th7uomnuzs1bd ... er.7z?dl=0

[Ozymandias]

Wow, twice as fast as the original. Can we get a compile of 15.7? (Initial commit). This is actually 15.8, BTW, Werner has already pointed to your link, in that thread.

[Dann Corbit]

So far as I know, those sources are not available.

[Ozymandias]

True, the earliest source is from August 7th. The Initial Commit, only has LICENSE and README.

[Jamal Bubker]

[quote="Dann Corbit"]Version with source and projects, along with correct icon and CPU type of Any(switches as appropriate) and X64 (hardwired to 64 bit)
Uses dotnet 4.0

https://www.dropbox.com/s/th7uomnuzs1bd ... er.7z?dl=0[/quote]

Thank you very much Dann !
So nice from you to have compiled a x64 binary

[Mike S.]

Surprisingly, Alfil 15.8 has just been disqualified from TCEC/S8, having been accused of using an internal book. I do not know which compile was running there, but your 64-bit version Dann, is so small that I do not consider it has a book inside. Also, it's behaviour does not seem to indicate that (but I did not take much time now to investigate).

Anyway, I have a feeling that a wrong decision may have been done. - In case you are interested, your word certainly has value.

[esacosta]

I was talking with Anton Mihailov and he says me that Alfil has doing that Perdone crash. I don´t know why. It seems that Alfil does not working with the option OwnBook = NO

Alfil use an internal book but you can disabled with OwnBook = NO. Maybe this version has something wrong, I need to review. But now I´m working to include tablebases to Alfil (i´m using new Gaviota tablebases)

[Dann Corbit]

Surprisingly, Alfil 15.8 has just been disqualified from TCEC/S8, having been accused of using an internal book. I do not know which compile was running there, but your 64-bit version Dann, is so small that I do not consider it has a book inside. Also, it's behaviour does not seem to indicate that (but I did not take much time now to investigate).

Anyway, I have a feeling that a wrong decision may have been done. - In case you are interested, your word certainly has value.

I guess it was the C++ version that was used.
Personally, I don't see anything wrong with storing 1.d4 in your program.
There is no internal book in the dotnet version. Maybe one exists in the C++ version.

There was also some question about relationship to stockfish engine.
The lineage is pretty clear with the dotnet version, but since it is open source and GPL like the original I do not think there is any license violation. At least with the dotnet version, there is no wrongdoing here that I can see.[*]

I did not do any check for innovation in the sources. So I do not know the degree of originality.

Besides the translation to dotnet, there is also a translation of data structures and variable names to another language (Looks like Spanish to me, but Portuguese is similar enough it might fool me). Clearly, a lot of work went into it.

I have no idea how similar the C++ version is to Stockfish.

[*] Mention of Stockfish would certainly be sensible in the release notes.

[Branko Radovanovic]

Surprisingly, Alfil 15.8 has just been disqualified from TCEC/S8, having been accused of using an internal book. I do not know which compile was running there, but your 64-bit version Dann, is so small that I do not consider it has a book inside. Also, it's behaviour does not seem to indicate that (but I did not take much time now to investigate).

Anyway, I have a feeling that a wrong decision may have been done. - In case you are interested, your word certainly has value.

Absolutely a decision contrary to TCEC rules, which say:

In the case of a serious, play-limiting bug (like crashing or interface communication problems) not discovered during the pre-Season testing, the engine can be updated once per Stage to fix this/these bug/bugs only. If this update still doesn't fix the problem(s) or if there is no update available, the engine might have the number of cores reduced, have the hash size reduced or have the tablebase access disabled - these changes will remain for the rest of the Stage.

Chessdom press release said:

Even though Alfil had its book switched off in the UCI, it still continued to use it.

But that's the very definition of a "serious, play-limiting bug" - a previously undiscovered bug that directly caused a violation of the tournament rules and thereby instantly forfeited the game (similar to e.g. playing an illegal move). So, the correct decision would have been to adjudicate the game as a win for Pedone and allow Alfil to be updated for the next game. Frankly, I'm amazed to see nobody is raising a stink about this.

[michiguel]

Surprisingly, Alfil 15.8 has just been disqualified from TCEC/S8, having been accused of using an internal book. I do not know which compile was running there, but your 64-bit version Dann, is so small that I do not consider it has a book inside. Also, it's behaviour does not seem to indicate that (but I did not take much time now to investigate).

Anyway, I have a feeling that a wrong decision may have been done. - In case you are interested, your word certainly has value.

Absolutely a decision contrary to TCEC rules, which say:

In the case of a serious, play-limiting bug (like crashing or interface communication problems) not discovered during the pre-Season testing, the engine can be updated once per Stage to fix this/these bug/bugs only. If this update still doesn't fix the problem(s) or if there is no update available, the engine might have the number of cores reduced, have the hash size reduced or have the tablebase access disabled - these changes will remain for the rest of the Stage.

Chessdom press release said:

Even though Alfil had its book switched off in the UCI, it still continued to use it.

But that's the very definition of a "serious, play-limiting bug" - a previously undiscovered bug that directly caused a violation of the tournament rules and thereby instantly forfeited the game (similar to e.g. playing an illegal move). So, the correct decision would have been to adjudicate the game as a win for Pedone and allow Alfil to be updated for the next game. Frankly, I'm amazed to see nobody is raising a stink about this.

Alfil was using an internal opening book because it failed to recognized properly the UCI command ownbook to be "false". It has code to switch it off, but it does not. That is a bug, not cheating. We do not know what was the communication between the organizers and the author, which would be important to express an opinion. But, of course, you do have a point.

There are other allegations about being a SF derivative, but it does not seem it has played a role in the decision.

Miguel