Bored with orthodox Chess?

[hgm]

Well, the packing was only done on continuation moves with the same piece. But it was indeed a bad idea, and I already abandoned it.

[Daniel Shawul]

Hi pawel
I just blindly tried to use the whole thing just for HG's test. I know there are a whole lot of things to think about null move being one of them. Should I use it in reversi & go where null move is legal, what about checkers ? I haven't thought about it yet and this is part of the fun I don't want to be spoiled for me I wanted to remove all the hacks first to allow same code to be used for all.
cheers

[Daniel Shawul]

Sorry rep-draw is my fault. It is Nebiyu who is sending all those "draw" and "stalemate" messages to winboard.

[hgm]

OK, new version at http://hgm.nubati.net/WinBoard-4.5.beta.zip .

This one should give (comma-separated) SAN multi-moves, and can actually read them back from a PGN file.

You can enter partial moves with the mouse by keeping the control key pressed.

It knows variant dark, but it only implements the partial hiding of the board in Machine White or Machine Black mode. (Difficult to test without an engine, of course, but I can duplicate variant normal in the fmax.ini file and rename it to 'dark', so that Fairy-Maxthinks it can play dark chess. The engine->GUI communication is not implemented yet, so it will indeed be playing normal Chess, but the Human user plays semi-blindfold.)

On the downside

The animation of multi-moves is still a mess, and leaves pieces on the display where they are not.

In variant dark there is no suppression of highlights and text move, so it is not only Fairy-Max that cheats, but the user as well.

No one-click drop moves yet.

[Daniel Shawul]

f6e5 {+1.53/24} 2. c3d4 {-1.29/25} e5c3 3. d2b4 {-1.29/26} d6c5 {+1.31/23} 4. b4d6 c7e5 {+1.32/23} 5. b2c3 {-1.25/23} b8c7 {+1.08/25} 6. a1b2 {-1.05/25} e5d4 {+0.92/24} 7. c3e5 {-0.92/27} c7d6 {+0.89/28} 8. e5c7 e7d6 {+0.80/25} 9. b2c3 {-0.74/25} g7f6 {+0.77/24} 10. c1b2 {-0.73/26} f8e7 {+0.72/25} 11. g3h4 {-0.53/28} h6g5 {+0.45/28} 12. f2g3 {-0.29/31} f6e5 {+0.23/28} 13. h4f6 e7g5 14. c3d4 {+0.13/28} e5c3,c3a1=Q 15. e3f4 {+0.17/26} g5e3 16. e1d2 {+0.19/25} e3c1=Q 17. g3f4 {+0.12/25} d6e5 {-0.09/26} 18. f4d6 Qab2 {-0.09/26} 19. h2g3 {+0.04/26} Qcd2 {-0.02/24} 20. a3b4 {+0.06/27} Qdc3 {+0.00/28} 21. b4c5 {+0.00/28} b6d4 22. g1h2 {+0.00/28} d8b6 23. g3h4 {+0.00/31} b6c5 {+0.00/31} 24. d6e7 {+0.00/30} a7b6 {+0.00/32} 25. h2g3 {+0.00/31} d4e3 {+0.00/31} 26. e7f8=Q {+0.00/28} Qcb4 {+0.00/30} 27. Qg7 {+0.00/44} h8f6 28. h4g5 {+0.00/58} f6h4,h4f2#

Yes it shows multi-moves in the PGN and # sign is only at the end.

[PK]

hi,

just uploaded Oberon 0.03 - it tests for king promotion legality and one search bug has been fixed. I don;t expect it to be significantly stronger, still in micro-Max ballpsrk.

[Richard Allbert]

Hi HG,

Link to the corrected Catalyst is here:

http://www.bluefever.net/enginepage.html

I finally have a website...

[hgm]

Somehow this corrected version is not an improvement. It freezes very often with white, usually after black exposes (or keeps exposed) one of his Kings to capture. That means it either genuinely crasheds, or it refuses the move without sending the proper "Illegal move" message to WB (which then would forfeit it for "False illegal move claim", but at least it would save you the time to wait it out).

It also does illegal moves. one time it left both its Kings in check, one time it stepped into check with its second King, while the first one was already in check. (But but there by Catalyst itself, so it does not always object to that.)

A very strange one is where it tries to promote a Hoplite to 's'. What the heck is an 's' supposed to be?

As a result of all these troubles, its score against Oberon dropped from 25% to 6%...

[Richard Allbert]

s = SpartanKing

Should I use k?

As for the other problems, I'm sorry about that, I'll have a look later today!

Ciao

Richard

[hgm]

Indeed, k = King.