A Glimmer Of Hope- Rds. 33 thru 36 & Final Standings!

[geots]

A GLIMMER OF HOPE

Intel Pentium 4
Single Processor- 2.8ghz
128mb hash
3 & 4 man tablebases
Ponder & Learning = OFF
Perfect 12.32.ctg book for all (w/12 move CCRL Limit)
40 moves in 63 minutes repeating (adapted to CCRL 40/40)
All engines = 32 bit
RR = 4 cycles
36 rounds = 180 games




Round 33

ChessAlex 2.0r4 v BlackBishop 0.47 draw
Uralochka 1.1b v Atak 6.0 (0-1)
Lime 62 v Firefly 2.5.0 (1-0)
Bearded 06-29 v Mustang 4.97 draw
BigLion 2.23x v Clarabit 1.00 draw


Round 34

BlackBishop 0.47 v BigLion 2.23x (1-0)
Clarabit 1.00 v Bearded 06-29 (1-0)
Mustang 4.97 v Lime 62 (0-1)
Firefly 2.5.0 v Uralochka 1.1b (0-1)
Atak 6.0 v ChessAlex 2.0r4 (0-1)


Round 35

Atak 6.0 v BlackBishop 0.47 (0-1)
ChessAlex 2.0r4 v Firefly 2.5.0 (0-1)
Uralochka 1.1b v Mustang 4.97 (1-0)
Lime 62 v Clarabit 1.00 (0-1)
Bearded 06-29 v BigLion 2.23x draw


Round 36

BlackBishop 0.47 v Bearded 06-29 (0-1)
BigLion 2.23x v Lime 62 draw
Clarabit 1.00 v Uralochka 1.1b draw
Mustang 4.97 v ChessAlex 2.0r4 draw
Firefly 2.5.0 v Atak 6.0 (1-0)


Final Standings

22.0- ChessAlex 2.0r4
21.0- BlackBishop 0.47
20.5- Bearded 06-29
19.5- Clarabit 1.00
19.5- BigLion 2.23x
18.5- Uralochka 1.1b
16.5- Mustang 4.97
16.0- Atak 6.0
14.0- Lime 62
12.5- Firefly 2.5.0


Congrats to ChessAlex and "thank you" to all the programmers whose work made this possible.


All rounds can be followed and all PGNs downloaded from:
http://kirill-kryukov.com/chess/discuss ... f=7&t=3408

[Dr.Wael Deeb]

Hi George,
Thanks for the results....

In my rating list at long time controls with more than 200 games played,the gap between the 1.xx series and the latest ChessAlex is around 200 Elo,a huge improvement

ChessAlex 2.0 r4 2035 236 games played
.
.
ChessAlex 1.4r3 1827 204 games played

[Graham Banks]

Thanks for running this tournament George.
The biggest surprise for me was the poor performance of Lime, but these things can happen in one off tournaments as most know.
ChessAlex has caught my eye!

[Matthias Gemuh]

Final Standings

22.0- ChessAlex 2.0r4
21.0- BlackBishop 0.47
20.5- Bearded 06-29
19.5- Clarabit 1.00
19.5- BigLion 2.23x
18.5- Uralochka 1.1b
16.5- Mustang 4.97
16.0- Atak 6.0
14.0- Lime 62
12.5- Firefly 2.5.0

Thanks, George.
I will not weep because I know one of BigLion's problems.
Maybe I will fix that soon.

Matthias.

[geots]

Final Standings

22.0- ChessAlex 2.0r4
21.0- BlackBishop 0.47
20.5- Bearded 06-29
19.5- Clarabit 1.00
19.5- BigLion 2.23x
18.5- Uralochka 1.1b
16.5- Mustang 4.97
16.0- Atak 6.0
14.0- Lime 62
12.5- Firefly 2.5.0

Thanks, George.
I will not weep because I know one of BigLion's problems.
Maybe I will fix that soon.

Matthias.

Short of one or two problems it has, it plays a doggone good game of chess. It certainly has potential.

Best,

[Tony Thomas]

Do not say doggone george, goat prefers the term goatgone. A nice tournament, rather unpredictable ending, I would have expected Blackbishop to win this one..

[Richard Allbert]

Thanks for running this tournament George.
The biggest surprise for me was the poor performance of Lime, but these things can happen in one off tournaments as most know.
ChessAlex has caught my eye!

I agree... although v62 has a massive hash bug which can cause piece losing blunders, especially in the endgame - at the very least, it chooses poor lines.

The test v63 that some people use also has a bug, just not as bad.

At the moment, I have a v64 on my computer that doesn't seem to have the bug, but it is slower than the previous versions - does better against test positions an worse in matches.

As I wrote in the WBEC forum, for the last few months I've been involved in something else (not programming related), but I do hope to have a new version out at some point.

As crazy as it sounds, I'm not finding it easy to improve on 62 / 63 enough to justify a new release.

Thanks for using Lime, though.

Richard