The importance of "Dynamism" in Komodo

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
cma6
Posts: 149
Joined: Thu May 29, 2014 3:58 pm

Default dynamism

Post by cma6 » Fri Apr 01, 2016 12:18 am

LK:
This looks like a big reversal on your part as regards default Dynamism. Your original preliminary suggestion for Dynamism for deep analysis (or at very long tc) was "80".
Are you now suggesting "110" instead of "80"?
Thanks,
CMA

lkaufman
Posts: 3880
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Default dynamism

Post by lkaufman » Fri Apr 01, 2016 3:12 am

cma6 wrote:LK:
This looks like a big reversal on your part as regards default Dynamism. Your original preliminary suggestion for Dynamism for deep analysis (or at very long tc) was "80".
Are you now suggesting "110" instead of "80"?
Thanks,
CMA
I'm saying that 80 produces more realistic evals and might be better for preparation for human games, but 110 produces better results in engine vs engine games. It seems that an unrealistic eval helps in engine vs engine play. We can speculate as to why this is so, but it's pretty clear to me that it is a real phenomenon, not a statistical fluke.
Komodo rules!

lkaufman
Posts: 3880
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: The importance of "Dynamism" in Komodo

Post by lkaufman » Fri Apr 01, 2016 3:15 am

Nordlandia wrote:Question related to Komodo's time usage heuristics.

Do ponder make experimenting with "Time Usage" parameter more appealing? assuming two computer face each others.

I was thinking about threshold range of -3 to +3 depening on advice.

Ponder might compensate for mild time usage adjustment, that was my idea.

We compensate for ponder, so there is no particular reason to think that Time Usage should be different for Ponder on games than for Ponder off, although of course our compensation may be wrong either in general or against specific opponents. It is also possible that the proper setting depends on the time control. These are all possible, but I think that experimenting with Dynamism and/or King Safety is more likely to bear fruit.
Komodo rules!

shrapnel
Posts: 1294
Joined: Fri Nov 02, 2012 8:43 am
Location: New Delhi, India

Re: Default dynamism

Post by shrapnel » Sat Apr 02, 2016 4:44 am

lkaufman wrote:It seems that an unrealistic eval helps in engine vs engine play. We can speculate as to why this is so, but it's pretty clear to me that it is a real phenomenon, not a statistical fluke.
+1.
This is indeed very much so, illogical as it may seem.
I suppose this is why Chess is called both an Art and a Science.
Everything about it cannot be explained in terms of hard, cold Logic.
Sometimes, setting more aggressive settings gives better results against even strong Engines, though the aggressive settings may not be ideally the best.
Maybe this is where the "Art" part of Chess comes in ; like in War, sometimes Attack is the best Defense !
i7 5960X @ 4.1 Ghz, 64 GB G.Skill RipJaws RAM, Twin Asus ROG Strix OC 11 GB Geforce 2080 Tis

leavenfish
Posts: 282
Joined: Mon Sep 02, 2013 6:23 am

Re: Default dynamism

Post by leavenfish » Sat Apr 02, 2016 4:08 pm

Whatever you do for Komodo 10 - PLEASE keep this feature! One might be good for 'ratings' against other programs, but most everyone else uses the engine for their own personal evaluations of positions for items like their own personal opening repertoire.

Hurnavich
Posts: 722
Joined: Wed Jun 19, 2013 6:14 pm

Re: Default dynamism

Post by Hurnavich » Sat Apr 02, 2016 5:34 pm

+1
"May your next game be your best"

User avatar
Leto
Posts: 2029
Joined: Thu May 04, 2006 1:40 am
Location: Dune

Re: The importance of "Dynamism" in Komodo

Post by Leto » Mon Apr 11, 2016 9:24 pm

Laskos wrote:
Laskos wrote:I was curious what does this mean:
Larry Kaufman wrote:The other new option, "Dynamism", gives the user full control over the style of Komodo; a low setting makes it prioritize static advantages like material and pawn structure, while a high setting prioritizes short-term ("dynamic") advantages like mobility and threats to the king. Although the default (100) is believed to be best for actual play, a lower value around 80 seems to produce more realistic evaluations in most positions so may be better for opening analysis.
I decided to use testsuites of different character to check for this setting. STS 1-15 - mostly positional in character, and Arasan 18 - mostly tactical.

STS: 5s/move

Code: Select all

Dynamism  Solved

 80        1346            
100        1367    ------
120        1341
The optimum is at the value of Dynamism 100.

Arasan: 5s/move

Code: Select all

Dynamism  Solved
100         126
140         155  
180         159
190         174    ------
200         171
The optimum is the value of Dynamism 190. The improvement compared to the default number of solved positions is huge, in fact we will see that combined with "King Safety", "Dynamism = 190" makes a tactical monster of Komodo.

So, in tactical problems, high values of Dynamism are favored. Maybe Larry's value of 80 is applicable to very quiet, positional problems?

Then, I also tested several hard tactical suites, and it is possible that Komodo 9.42 Dynamism=190; King Safety=160 (or similar values, try them) is the new King of tactics, maybe at least on a par with Houdini 4 Tactical, way above SF Matefinder. The testsuites were Arasan 18, HardTalkchess2015, DieRybka with about 5s/move.
Komodo D 190 KS 160 beats Houdini 4 Tactical in new Arasan 19 beta suite:
5s/move, 4 cores:
K: 178/250
H: 172/250

My impression is that by varying "Dynamism" from say 60 to 190, Komodo became both the strongest positional engine and the strongest tactical engine. Great feature!
I tested your settings with 12 cores with the Arasan 18 test suite and at 30s limit it seems with more time to think this setting can't keep up with Houdini 4 Tactical.

With your settings at 30 seconds limit and 12 cores I got 208/250 83.2%, this is just slightly better than Komodo 9.3 (202/250 80.7%, but significantly better than Komodo 9.42 (186/250 74.4%).

Houdini 4 Tactical still has the highest score in my testing with Arasan 18 testsuite at the 30s limit, with 240/250 (96%).

Post Reply