Komodo MCTS

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

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
peter
Posts: 1760
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: Komodo MCTS

Post by peter » Sat Jun 15, 2019 10:52 am

marsell wrote:
Sat Jun 15, 2019 9:36 am
The engine BrainLearn from Kelly Kinyama produces significantly more files than 8. Nearly 900
What am I doing wrong??
You're not doing anything wrong. If you read my post carefully, I wrote, 8 files per single different opening- line. As much different openings, as much .bin- files (except experience.bin remaining a single one) x7 + 1 experience.bin.
If you stick to a single line of opening- moves of >7 moves length, there will be 8 files only but constantly growing with the amount of games played with the same one opening.

That's my way of doing it, having separate folders of .bin- files for each and every single opening line (e.g. corr.- game) of interest of mine.
Peter.

marsell
Posts: 19
Joined: Tue Feb 07, 2012 10:14 am

Re: Komodo MCTS

Post by marsell » Sat Jun 15, 2019 11:15 am

okay, thanks for the clarification.

sound67
Posts: 4
Joined: Tue May 14, 2019 10:39 am
Full name: Thomas Muething

Re: Komodo MCTS

Post by sound67 » Tue Jun 18, 2019 10:41 am

I played an engine torunament between Komodo 13, Komodo 13 MCTS and Stockfish 10 - on a P5-8400, 2GB Hash, 5 Cores, 40/2h+20/1h+30 mode with the 2017perfect book. 90 games, took almost a month.

Komodo 13 MCTS finished last. It was beaten by its own "regular" version (+0, -5, =25) as well as by Stockfish (+0, -6, =24).

Regular Komodo 13 lost to Stockfish by +0, -9, =21.

From the individual games it looks as if neither version of the new Komodo has any hope to beat Stockfish, not even once.

User avatar
Ovyron
Posts: 2339
Joined: Tue Jul 03, 2007 2:30 am

Re: Komodo MCTS

Post by Ovyron » Tue Jun 18, 2019 9:13 pm

sound67 wrote:
Tue Jun 18, 2019 10:41 am
I played an engine torunament between Komodo 13, Komodo 13 MCTS and Stockfish 10 - on a P5-8400, 2GB Hash, 5 Cores, 40/2h+20/1h+30 mode with the 2017perfect book. 90 games, took almost a month.

Komodo 13 MCTS finished last. It was beaten by its own "regular" version (+0, -5, =25) as well as by Stockfish (+0, -6, =24).

Regular Komodo 13 lost to Stockfish by +0, -9, =21.

From the individual games it looks as if neither version of the new Komodo has any hope to beat Stockfish, not even once.
Now try the same match at MultiPV=4 (as some people use in analysis) :wink: :wink:
Make someone happy today.

Modern Times
Posts: 2370
Joined: Thu Jun 07, 2012 9:02 pm

Re: Komodo MCTS

Post by Modern Times » Tue Jun 18, 2019 9:49 pm

Ovyron wrote:
Tue Jun 18, 2019 9:13 pm
Now try the same match at MultiPV=4 (as some people use in analysis) :wink: :wink:
I've tried it, no noticeable difference in performance.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

sound67
Posts: 4
Joined: Tue May 14, 2019 10:39 am
Full name: Thomas Muething

Re: Komodo MCTS

Post by sound67 » Wed Jun 19, 2019 8:34 am

Ovyron wrote:
Tue Jun 18, 2019 9:13 pm
Now try the same match at MultiPV=4 (as some people use in analysis) :wink: :wink:
I used the maximum number of cores for the engines that my P5 processor allowed, MultiPV=5. Ponder=Off.

lkaufman
Posts: 3647
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Komodo MCTS

Post by lkaufman » Wed Jun 19, 2019 4:07 pm

sound67 wrote:
Wed Jun 19, 2019 8:34 am
Ovyron wrote:
Tue Jun 18, 2019 9:13 pm
Now try the same match at MultiPV=4 (as some people use in analysis) :wink: :wink:
I used the maximum number of cores for the engines that my P5 processor allowed, MultiPV=5. Ponder=Off.
Did you set MultiPV = 5 for ALL engines, not just Komodo? With that setting, Komodo 13.02 MCTS is MUCH stronger than normal Komodo, and about the same strength as Stockfish 10 (also using MultiPV). If you are getting results with Komodo MCTS below normal Komodo, something is wrong.
Komodo rules!

jp
Posts: 705
Joined: Mon Apr 23, 2018 5:54 am

Re: Komodo MCTS

Post by jp » Wed Jun 19, 2019 4:26 pm

lkaufman wrote:
Wed Jun 19, 2019 4:07 pm
With that setting, Komodo 13.02 MCTS is MUCH stronger than normal Komodo, and about the same strength as Stockfish 10 (also using MultiPV).
But why set SF multPV too, when that'll weaken it? What is the elo drop of SF with multiPV > 1?

lkaufman
Posts: 3647
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Komodo MCTS

Post by lkaufman » Wed Jun 19, 2019 4:35 pm

jp wrote:
Wed Jun 19, 2019 4:26 pm
lkaufman wrote:
Wed Jun 19, 2019 4:07 pm
With that setting, Komodo 13.02 MCTS is MUCH stronger than normal Komodo, and about the same strength as Stockfish 10 (also using MultiPV).
But why set SF multPV too, when that'll weaken it? What is the elo drop of SF with multiPV > 1?
The idea is to test the engines the way many users actually use them, with MultiPV = 4 or 5. The drop in strength in SF (or any A/B engine) with MCTS is nearly as bad as cutting the time by that ratio, so SF with MultiPV = 5 is like SF with one fifth of the time in strength, roughly speaking. Komodo MCTS doesn't gain in strength my displaying more moves, it just doesn't lose strength the way all A/B engines do. No one is suggesting that all testing be done this way, it's just useful for those who normally use MultiPV.
Komodo rules!

Modern Times
Posts: 2370
Joined: Thu Jun 07, 2012 9:02 pm

Re: Komodo MCTS

Post by Modern Times » Wed Jun 19, 2019 6:05 pm

lkaufman wrote:
Wed Jun 19, 2019 4:07 pm
sound67 wrote:
Wed Jun 19, 2019 8:34 am
Ovyron wrote:
Tue Jun 18, 2019 9:13 pm
Now try the same match at MultiPV=4 (as some people use in analysis) :wink: :wink:
I used the maximum number of cores for the engines that my P5 processor allowed, MultiPV=5. Ponder=Off.
Did you set MultiPV = 5 for ALL engines, not just Komodo?
Just Komodo.

Your point about MPV and analysis is well taken. Komodo MCTS is superb for analysis, and is even more compelling it you don't have a multi-core monster machine.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Post Reply