At the price of more lookup cost because 149 GB don't fit into the RAM. Plus that most of these lookups occur in branches that would have been cut off anyway. So it may well be that there is no real advantage; it might even actually weaken the engine.Nordlandia wrote: nevertheless 149 GB of perfect information 5-men don't have access to.
komodo 11.2.2
Moderator: Ras
-
- Posts: 2697
- Joined: Tue Aug 30, 2016 8:19 pm
- Full name: Rasmus Althoff
Re: komodo 11.2.2
-
- Posts: 2822
- Joined: Fri Sep 25, 2015 9:38 pm
- Location: Sortland, Norway
Re: komodo 11.2.2
Yeah, that's why i've switched to probing 5-men during games with 6-men adjudication. CuteChess 1.0.0 <Ras wrote:At the price of more lookup cost because 149 GB don't fit into the RAM. Plus that most of these lookups occur in branches that would have been cut off anyway. So it may well be that there is no real advantage; it might even actually weaken the engine.Nordlandia wrote: nevertheless 149 GB of perfect information 5-men don't have access to.
-
- Posts: 282
- Joined: Mon Sep 02, 2013 8:23 am
Re: komodo 11.2.2
Nevermind....user error.leavenfish wrote:So, I am analyzing a game position in Chessbase - Komodo 11.2.2 and I try to up the CPU 1 - 8 (have a i7 - 7600) and I can't get it to go to 4....stays at 1.
I do not think I've had this problem in earlier versions...
Any thoughts?

-
- Posts: 1339
- Joined: Fri Nov 02, 2012 9:43 am
- Location: New Delhi, India
Re: komodo 11.2.2
Yes. Certainly agrees with naked eye observations of Engine-Engine Matches. Practically speaking, results are more affected by strength of the Book, Engine and Hardware, in that order; Syzygy is a distant 4th, if at all.Ras wrote:At the price of more lookup cost because 149 GB don't fit into the RAM. Plus that most of these lookups occur in branches that would have been cut off anyway. So it may well be that there is no real advantage; it might even actually weaken the engine.
So, 5-men Syzygy is certainly more than enough in the real world of Engine-Engine Matches/Tournaments, except for the Purists.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis
-
- Posts: 626
- Joined: Sun May 13, 2007 9:55 pm
- Location: Bay Area, CA USA
- Full name: Mike Adams
Re: komodo 11.2.2
I think there is a diminishing rate of return with table base. And engines evaluates are so good they mostly play the endgame based on evaluate rules just fine. Throw in some 5 or less man table base your sitting tall on those rare endgames where a little more is needed. The chances of finding a situation where a draw would not be got or a win lost without 6 man must be rather low in my view.
-
- Posts: 1339
- Joined: Fri Nov 02, 2012 9:43 am
- Location: New Delhi, India
Re: komodo 11.2.2
+1.adams161 wrote:The chances of finding a situation where a draw would not be got or a win lost without 6 man must be rather low in my view.
Well put.
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis
-
- Posts: 2822
- Joined: Fri Sep 25, 2015 9:38 pm
- Location: Sortland, Norway
Re: komodo 11.2.2
Rule of thumb ->
6-men for analysis
5-men for engine games combined with 6-men adjudication [CuteChess 1.0.0]
6-men for analysis
5-men for engine games combined with 6-men adjudication [CuteChess 1.0.0]
-
- Posts: 3712
- Joined: Thu Jun 07, 2012 11:02 pm
Re: komodo 11.2.2
Agreed, although I currently only use 5-men for adjudication as I don't use CuteChess !Nordlandia wrote:Rule of thumb ->
6-men for analysis
5-men for engine games combined with 6-men adjudication [CuteChess 1.0.0]