CCRL Engine Authors' Tournament (will start October 1st)

[D Sceviour]

Thank you Guenther.

It does appear to be a timer error, which occurs in ChessGUI but it does not appear in Arena or Winboard. Without a debug file there is no way to tell for sure. In the meantime, it would be a good idea to withdraw Schooner 1.5.7 from ChessGUI tournaments because there are false results being turned in.

[Guenther]

Just for the record this problem does not appear in the games of Sergio under a different GUI (probably WB).

Either the tc is sent different in ChessGUI or it is a problem of Schooner which only appears in the ChessGUI.

I see our posts crossed now.

[D Sceviour]

Just for the record this problem does not appear in the games of Sergio under a different GUI (probably WB).

Either the tc is sent different in ChessGUI or it is a problem of Schooner which only appears in the ChessGUI.

I see our posts crossed now.

Sergio is using Arena, I am fairly sure.

[Modern Times]

Sergio is using Arena, I am fairly sure.

I think that is right, I've never seen him use anything else when submitting games.

[Sergio Martinez]

Yes, like Ray says I normally use Arena and in this tournament Im using Arena 3.51

[Graham Banks]

Thank you Guenther.

It does appear to be a timer error, which occurs in ChessGUI but it does not appear in Arena or Winboard. Without a debug file there is no way to tell for sure. In the meantime, it would be a good idea to withdraw Schooner 1.5.7 from ChessGUI tournaments because there are false results being turned in.

I am using Schooner as a winboard engine.
It certainly does have a time issue after the first time control as you've all noticed.
I can email debug files to you to look at.

[Graham Banks]

Thank you Guenther.

It does appear to be a timer error, which occurs in ChessGUI but it does not appear in Arena or Winboard. Without a debug file there is no way to tell for sure. In the meantime, it would be a good idea to withdraw Schooner 1.5.7 from ChessGUI tournaments because there are false results being turned in.

I am using Schooner as a winboard engine.
It certainly does have a time issue after the first time control as you've all noticed.
I can email debug files to you to look at.

I'll add that previous versions of Schooner didn't seem to have this issue under ChessGUI.

[Graham Banks]

Just trying out the Group E engines under ChessGUI.

Jumbo seems to run as it should.

Pigeon moves pretty much instantly every move.

Dorpsgek seems to run as it should.

Satana plays okay, but without displaying any information.

Neurone moves pretty much instantly every move.

Soberango moves instantly first move out of book, but is okay after that.

[D Sceviour]

Thank you Guenther.

It does appear to be a timer error, which occurs in ChessGUI but it does not appear in Arena or Winboard. Without a debug file there is no way to tell for sure. In the meantime, it would be a good idea to withdraw Schooner 1.5.7 from ChessGUI tournaments because there are false results being turned in.

I am using Schooner as a winboard engine.
It certainly does have a time issue after the first time control as you've all noticed.
I can email debug files to you to look at.

I'll add that previous versions of Schooner didn't seem to have this issue under ChessGUI.

I found a bug. According to Gabor Szots, CCRL testing requirements are that an engine must accept the GUI book moves using the CECP force command. The thread can be followed here:
http://talkchess.com/forum/viewtopic.ph ... 43&t=57311

ChessGUI does not follow this convention and the game move number is read instead from the CECP setboard command. This caused Schooner to read the 9th move as move number 1; because Schooner is waiting for the GUI book move as I thought was required.

This will require some timer code re-write and testing to get Schooner to comply with ChessGUI book bypass. The problem has been there for a while and explains the discrepancy with Sergio's Arena method and Gabor's requirement among other problems. It is unclear at this time if this is the only problem. Note: there is a minor bug in Schooner that asserts the time control as 40 moves per time control for tournaments, and never 32 moves for the first timer control. That will be fixed in the future. Blitz and incremental chess should work fine for any amount of time.

The good news is that Schooner will offer a 2400 program for those who use Arena or Winboard!

[hgm]

ChessGUI expects the engines to look at the full-move number of FENs to know when a new session starts?

That certainly breaks all of my engines!

I always assumed that all serious testing is done by feeding moves, and never by starting from positions. I often don't even bother to equip them with setboard.