Not logged inRybka Chess Community Forum
Up Topic Rybka Support & Discussion / Rybka Discussion / Hiarcs 11.1 humiliated by Miss Fish
- - By masomusic (**) Date 2007-06-26 13:33
   Engine               Score                                                  Ry                                                 Hi    S-B
1: Rybkav2.3.2a.w32     35,5/50 ·················································· 00==110====111111====1==11=1=11==1101=01=11111111=  514,75
2: Hiarcs11.1UCI [edit] 14,5/50 11==001====000000====0==00=0=00==0010=10=00000000= ··················································  514,75

50 games played / Tournament is finished
Name of the tournament: Arena tournament
Site/ Country: P4, Nederland
Level: Tournament 40/15 40/15 Rest:15
Hardware: Dual        Intel(R) Pentium(R) 4 CPU 3.00GHz  with 1.024 MB Memory
Operating system: Microsoft Windows XP Professional Service Pack 2 (Build 2600)
Attachment: Hiarcs11.1vsRybka2.3.2a.pgn (535k)
Attachment: Hiarcs11.1vsRybka2.3.2a.txt (823B)
Parent - - By Carl Bicknell (*****) Date 2007-06-26 15:43
71% - I think that's pretty typical. Did you notice a pattern in the games? From what I've heard it's mainly in the endgame that hiarcs gets pounded.
Parent - By masomusic (**) Date 2007-06-27 14:26
take a look at the attachments
Parent - - By masomusic (**) Date 2007-06-27 21:09
It's HIARC's positional overestimation that gets pounded. Unsound moves hat Rybka punishes well.
Parent - - By herakleitos (***) Date 2007-06-27 21:47
In my experience, Rybka is light-years ahead positionally, while Hiarcs is often more trenchant (and quicker) in more purlely tactical situations. I use them as complementary engines, and go back and forth. They cover each others' incompletenesses very effectively.
Parent - - By Carl Bicknell (*****) Date 2007-06-28 08:04
I do the same thing but with Junior instead of Hiarcs, I also find it works well.
Parent - By herakleitos (***) Date 2007-06-28 14:31
Interesting. I'll try this. Thanks.
Parent - - By Leto (***) Date 2007-06-28 15:14 Edited 2007-06-28 15:18
In my tactical testing Hiarcs 11.1 scores much higher than Junior, in fact if it wasn't for Rybka WinFinder Hiarcs would be the most tactical engine.  Another interesting fact is that Rybka 2.3.2 is more tactical than Junior 10!  Seeing as how Rybka is very positional but not very strong in tactics, I think the Rybka + Hiarcs combo would be stronger.  Perhaps even better than that combo would be Rybka + Rybka WinFinder 2.2. 

You can see my tactical testing results here, make sure to also see the second page where I include Rybka 2.3.2 in the list:
http://64.68.157.89/forum/viewtopic.php?t=12445&highlight=
Parent - - By Carl Bicknell (*****) Date 2007-06-28 16:22
In terms of finding winning combinations that's true. But I use Junior because it suggests moves that the other progs wont. Sometimes I look at a position and I think "I'd like to play Nxf7 here but Rybka doesn't suggest it so maybe it's no good. Time to switch to Junior" and sometimes Junior likes the move so I feel good about that.

For general play - Rybka
For tactics - Rybka Winfinder
For "attacking possibilities" - Junior
Parent - - By Harvey Williamson (*****) Date 2007-06-28 16:53
If you think Nxf7 might work you can of course force any engine to play it and then leave it to analyse for a few minutes you will soon know if it works.
Parent - By Carl Bicknell (*****) Date 2007-06-28 19:18
It's not a tactical thing. It's a preference. It's like whether or not to play g4 in some qgd positions. Some programs will like the thrust others wont. I tend to find that Junior typically likes those sort of moves. If even IT doesn't like it then I'd probably conclude it wasn't any good. But if Rybka, say, rejected it I wouldn't draw any conclusions.
Up Topic Rybka Support & Discussion / Rybka Discussion / Hiarcs 11.1 humiliated by Miss Fish

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill