Stockfish 2.3 update available

[Modern Times]

Hmm... the GCC SSE42 version just crashed under chessgui and chess960. I'll investigate further.

OK - looks like chess960 support is broken.

Under Shredder GUI, with the generic x64 compile and the GCCSSE42 compile. Stockfish crashes immediately a game starts. It is fine on normal chess.

[mcostalba]

Hmm... the GCC SSE42 version just crashed under chessgui and chess960. I'll investigate further.

OK - looks like chess960 support is broken.

Under Shredder GUI, with the generic x64 compile and the GCCSSE42 compile. Stockfish crashes immediately a game starts. It is fine on normal chess.

Could you please post the fen of the position that leads to a crash? Thanks.

[IWB]

No, I always use a 64 bit compile if available. In this case both 64bit compiles.
It did not even crossed my mind to use the 32bit in this case ... 32bit is so far off already

Bye
Ingo

[Modern Times]

Could you please post the fen of the position that leads to a crash? Thanks.

[Event "?"]
[Site "?"]
[Date "2005.07.07"]
[Round "?"]
[White "?"]
[Black "?"]
[Result "*"]
[SetUp "1"]
[FEN "rkqbnnbr/pppppppp/8/8/8/8/PPPPPPPP/RKQBNNBR w HAha -"]

*


I tried Stockfish on infinite analysis on a chess 960 position and it was fine (not the one above though). But when I start an engine match with the above it crashes

EDIT: Tried another position and the engine match started OK - so some are OK

[Modern Times]

If it helps:

Stockfish.VE_09 handles the position just fine.
http://talkchess.com/forum/viewtopic.ph ... kfish+ve09

I ran 1000 chess960 games with that and it ran perfectly.

[mcostalba]

[FEN "rkqbnnbr/pppppppp/8/8/8/8/PPPPPPPP/RKQBNNBR w HAha -"]

Thanks Ray, bug confirmed.

Could you pelase verify that with the following change to file movegen.cpp everything works?

Code: Select all

--- a/src/movegen.cpp
+++ b/src/movegen.cpp
@@ -46,7 +46,7 @@ namespace {
 
     assert(!pos.in_check());
 
-    for (Square s = kto; s != kfrom; s += (Square)(Side == KING_SIDE ? -1 : 1))
+    for (Square s = kto; s != kfrom; s += (Square)(kto > kfrom ? -1 : 1))
         if (pos.attackers_to(s) & enemies)
             return mlist;

Thanks.

[Modern Times]

I don't know, Jim will have to recompile, I can't do that myself.

[mcostalba]

I don't know, Jim will have to recompile, I can't do that myself.

I'd prefer to wait some days before to ask Jim to push another fix release.

If you send me a pm with an email address I can send to you directly a version compiled by me (not fast as Jim but enough to confirm bug is fixed).

[mcostalba]

If you send me a pm with an email address I can send to you directly a version compiled by me (not fast as Jim but enough to confirm bug is fixed).

P.S: For people testing in normal chess there is no need to wait, go with the fastest version and have fun

[Laskos]

If you send me a pm with an email address I can send to you directly a version compiled by me (not fast as Jim but enough to confirm bug is fixed).

P.S: For people testing in normal chess there is no need to wait, go with the fastest version and have fun

Non SSE compiles, ultra-bullet 2.5s + 0.04s games

Code: Select all

    Program                            Score       %     Elo    +   -    Draws

  1 Stockfish 2.3 JA 64bit         : 3227.5/6000  53.8   3213    7   7   42.9 %
  2 Stockfish 2.2.2 JA             : 2772.5/6000  46.2   3187    7   7   42.9 %

26 +/-7 Elo points improvement at this TC, probably 10-15 Elo points improvement at longer TC.

Kai