OpenChess

Critter 1.4 wins the December, 2011 chess engine release!

Critter 1.4 is definitely a major improvement over previous version 1.2 (1.3 does not exist). The results are pretty good from my own private testing, and as can be seen on the rating lists. Congrats to Richard.

Komodo 4 comes in second. Lost points due to pre-release over-hyping, and its authors continuous bashing of Critter, even after the author sent them his source code out of generosity.


Other contenders [ordered by date of release]:

GreKo
Stockfish
Quazar
DoubleCheck
Rodent
EnginMax
Supra
Tornado
Macromix
Chiron
Deuterium
BagaturChess
IvanHoe
Sjaak
Sting SF
Bearded
Murka
Arasan
Mediocre
ProDeo

kingliveson wrote:and its authors continuous bashing of Critter

When has Don bashed Critter?

Uly wrote:

kingliveson wrote:and its authors continuous bashing of Critter

When has Don bashed Critter?

Komodo version 4
uci
id name Komodo64 SSE Version 4
id author Don Dailey, Larry Kaufman
option name Ponder type check default true
option name Hash type spin default 64 min 0 max 4096
option name OwnBook type check default true
option name MultiPV type spin default 1 min 1 max 99
option name Book File type string default komodo.bin
option name Clear Hash type button
option name Use Lmr type check default true
option name Null Move pruning type check default true
option name Deterministic type check default false
option name Best Book Line type check default true
option name Strength type spin default 15 min 0 max 15
option name Sudden Divisor type spin default 40 min 10 max 80
option name Fischer Divisor type spin default 30 min 1 max 80
option name move overhead milliseconds type spin default 20 min 0 max 50000
option name Drawscore type spin default -5 min -50 max 50
uciok


Is Larry Kaufman Komodo's author? Yes. I am sure you are aware without me going into details of the dog-whistle campaign to delegitimizing Critter. They should get back to improving their engine and cease the unproductive rhetoric.

Komodo only has 2 authors, by saying "its authors continuous bashing of Critter" you make it sound as if both Larry and Don were continuously bashing Critter. Which is unfair if Don has not only not bashed Critter, but hasn't done it "continuously". Otherwise you'd have only referred to Larry.

As I don't visit some other fora, I wouldn't know if Don continuously bashes Critter or not, thus the question.

If Dr. Dailey is not complacent, then he ought to advice his co-author to put an end to it.

Uly wrote:Komodo only has 2 authors, by saying "its authors continuous bashing of Critter" you make it sound as if both Larry and Don were continuously bashing Critter. Which is unfair if Don has not only not bashed Critter, but hasn't done it "continuously". Otherwise you'd have only referred to Larry.

As I don't visit some other fora, I wouldn't know if Don continuously bashes Critter or not, thus the question.

Don D. never wrote anything negative at all about Critter in fora I visit. Larry K. did, but I am sure he did not meant any harm. He just sincerely stated what he believes.
(Komodo does not need to be stronger, it is already #1.
Houdini is even stronger? - doesn't count, it is a 'dirty' clone after all
Ippolit? - unscrupulous hacked Rybka
Critter? - it is (albeit most original [sic]) a member of IPPO family
Rybka4? - already banned from ICGA. If it would not - R3 evaluation is my work (LK) so what?
Stockfish? - some opensource guys trying to better us, nevertheless our search is based on it
)

Richard Vida wrote: Stockfish? - some opensource guys trying to better us, nevertheless our search is based on it

Of course K3 search is _not_ based on SF, contrary to what LK said. On the Rybka forum LK wrote that Komodo's search is more similar to SF than Ippolits, but from the programmer perspective I can not agree with that. Either Larry does not have access to (or can not understand) Don's code, but K3 search shares a _lot_ with R3/Ippo search. [I don't know about K4 because I don't want to pay for it - I naively thought they will send me a copy in exchange for my sources ].

Disclaimer: Above was not an accusation, K3 does not contain foreign code. Don seems to be quite obsessed with his own dogmas that nobody would copy in 21th century. I think Komodo will stay original in some sense no matter what.
[copy+make = undo for free, but he needs 1-5 branches just to determine which piece is moving / captured //EDIT: that is with 64bit version... in 32bits it is much worse]

Well, congrats for winning Engine Of The Month Richard, and I think that for interactive analysis of corr games (not for people that just let their engine sitting in the same position for hours, or those that only let an automated algorithm "interact" like in IDEA) you already have the best engine out there

I think it is a good idea that, at beginning/middle of every month, from Computer-Chess Wiki, we pick an engine of the month from previous month. The score will be based on Elo (versus previous version), new features, bug fixes, user experience, opinions, etc.