AGAIN- A TRIPLE TIE At the Top Concludes GROUP 10!

Discussion of computer chess matches and engine tournaments.

Moderator: Ras

User avatar
geots
Posts: 4790
Joined: Sat Mar 11, 2006 12:42 am

AGAIN- A TRIPLE TIE At the Top Concludes GROUP 10!

Post by geots »

The results are in- and the best I can say is it was one hell of a fight. And "it is what it is!"



Intel i5 w/4TCs
Fritz 11 gui/Fritz 13 gui
1CPU/32-bit or 64-bit where available
128MB hash
Bases=NONE [No egtb & no egbb]
Ponder_Learning=OFF
Perfect 2012b.ctg w/12-move limit
40/21 Repeating (Benched to adapt to CCRL 40/40)
RR with 2 cycles





Group 10
-----------



Round 26
-----------

Zeus 1.27 v AICE 0.99.2 (1-0)
Arion 1.7 v Asterisk 0.6 (1-0)
Mephisto Gideon Pro v ECE 12.01 (1-0)
Flux 2.2 v Mediocre 0.5 (draw)
Plisk 0.2.7_d x64 v Waxman 2011 (1-0)
TJchess 1.1U x64 v KnightX 1.92 (draw)
Sungorus 1.4 x64 v Averno 0.81 (1-0)
Popochin 2.9 v Sjakk 1.1.9 (0-1)

Round 27
-----------

Popochin 2.9 v Zeus 1.27 (1-0)
Sjakk 1.1.9 v Sungorus 1.4 x64 (1-0)
Averno 0.81 v TJchess 1.1U x64 (0-1)
KnightX 1.92 v Plisk 0.2.7_d x64 (1-0)
Waxman 2011 v Flux 2.2 (0-1)
Mediocre 0.5 v Mephisto Gideon Pro (draw)
ECE 12.01 v Arion 1.7 (draw)
Asterisk 0.6 v AICE 0.99.2 (1-0)

Round 28
-----------

Zeus 1.27 v Asterisk 0.6 (0-1)
AICE 0.99.2 v ECE 12.01 (draw)
Arion 1.7 v Mediocre 0.5 (1-0)
Mephisto Gideon Pro v Waxman 2011 (1-0)
Flux 2.2 v KnightX 1.92 (1-0)
Plisk 0.2.7_d x64 v Averno 0.81 (1-0)
TJchess 1.1U x64 v Sjakk 1.1.9 (1-0)
Sungorus 1.4 x64 v Popochin 2.9 (0-1)

Round 29
-----------

Sungorus 1.4 x64 v Zeus 1.27 (draw)
Popochin 2.9 v TJchess 1.1U x64 (1-0)
Sjakk 1.1.9 v Plisk 0.2.7_d x64 (0-1)
Averno 0.81 v Flux 2.2 (1-0)
KnightX 1.92 v Mephisto Gideon Pro (0-1)
Waxman 2011 v Arion 1.7 (0-1)
Mediocre 0.5 v AICE 0.99.2 (draw)
ECE 12.01 v Asterisk 0.6 (draw)

Round 30
-----------

Zeus 1.27 v ECE 12.01 (0-1)
Asterisk 0.6 v Mediocre 0.5 (1-0)
AICE 0.99.2 v Waxman 2011 (draw)
Arion 1.7 v KnightX 1.92 (1-0)
Mephisto Gideon Pro v Averno 0.81 (draw)
Flux 2.2 v Sjakk 1.1.9 (0-1)
Plisk 0.2.7_d x64 v Popochin 2.9 (1-0)
TJchess 1.1U x64 v Sungorus 1.4 x64 (1-0)

Code: Select all


Final Standings
----------------
1.Flux 2.2             18.5---275.50**
2.Mephisto Gideon Pro  18.5---262.75**
3.Arion 1.7            18.5---259.00
4.TJchess 1.1U x64     18.0
5.Sjakk 1.1.9          17.5
6.Plisk 0.2.7_d x64    17.0
7.Asterisk 0.6         17.0
8.Mediocre 0.5         15.5
9.ECE 12.01            14.5
10.AICE 0.99.2         14.5
11.Averno 0.81         14.0
12.Popochin 2.9        12.0
13.Waxman 2011         12.0
14.Zeus 1.27           11.5
15.KnightX 1.92        10.5
16.Sungorus 1.4 x64    10.5


Of course thanks to all these engines- and congrats especially to the top 3- Flux, Mephisto and Arion. It is actually a shame that one of these 3 have to go- but that's the way things get done.

So down the road, we will again see Flux and Mephisto- as they see a whole lot more of each other.

And:

1.Flux 2.2

2.Mephisto Gideon Pro


Best,

gts
User avatar
Rebel
Posts: 7382
Joined: Thu Aug 18, 2011 12:04 pm
Full name: Ed Schröder

Re: AGAIN- A TRIPLE TIE At the Top Concludes GROUP 10!

Post by Rebel »

George, just for the record what hash table size did you use for Gideon ?
User avatar
geots
Posts: 4790
Joined: Sat Mar 11, 2006 12:42 am

Re: AGAIN- A TRIPLE TIE At the Top Concludes GROUP 10!

Post by geots »

Rebel wrote:George, just for the record what hash table size did you use for Gideon ?

Ed, I used 128MB, but as I think I told you, I had to use- for whatever reason- "w6", which is actually for 256MB, to get it to show 128MB in your param.txt and also in Task Manager. So there is no doubt that it took "w6" instead of the usual "w4" to get 128MB. The first time I have ever run up on this. Never happened with ProDeo. But it did not bother me, because I never just set something and assume anything. I always double check.

Now I ran this tourn. on my Intel i5. If I had run it on XP, maybe "w4" would have given me 128MB. Who knows. But you did well.


Best,

george
User avatar
Rebel
Posts: 7382
Joined: Thu Aug 18, 2011 12:04 pm
Full name: Ed Schröder

Re: AGAIN- A TRIPLE TIE At the Top Concludes GROUP 10!

Post by Rebel »

geots wrote:
Rebel wrote:George, just for the record what hash table size did you use for Gideon ?

Ed, I used 128MB, but as I think I told you, I had to use- for whatever reason- "w6", which is actually for 256MB, to get it to show 128MB in your param.txt and also in Task Manager. So there is no doubt that it took "w6" instead of the usual "w4" to get 128MB. The first time I have ever run up on this. Never happened with ProDeo. But it did not bother me, because I never just set something and assume anything. I always double check.

Now I ran this tourn. on my Intel i5. If I had run it on XP, maybe "w4" would have given me 128MB. Who knows. But you did well.


Best,

george
Hi George,

I have been using a different set (w0-w9) for Gideon's hash tables because I wanted to emulate 1993 as close as possible and the default was set to 2Mb. If however you ran the tournament with 128Mb then that adds some significant strength and makes its good result more understandable for me.

Keep up the good work, must be an exhausting experiment :wink: