No recalibration (reference and current list are compared as they are) Recalibrate reference list to current one using selected engines only Recalibrate reference list to current one using all common engines Recalibrate current list to reference using selected engines only Recalibrate current list to reference using all common engines.
TopicThe Rybka Lounge / Computer Chess / Deep Sjeng 3 update
Over on Deep Sjeng's site there is reference to it suporting tablebases *new for 2008* so the blurb says.
Anyone recieved a free update?
Anyone recieved a free update?
All paying customers should receive an email with a download link and information about the new version somewhere in the next 24 hrs.
exciting, I love updates!
Do you have an estimated elo improvement yet?
Do you have an estimated elo improvement yet?
It will be fairly small and VERY dependent on the testing conditions, because the time usage changes are about half of it.
Do not expect too much. I'm mostly releasing this because so many people asked for endgame database support, and I guess also that people who bought Sjeng want the exact 'World Champion' version. I know I would :)
Do not expect too much. I'm mostly releasing this because so many people asked for endgame database support, and I guess also that people who bought Sjeng want the exact 'World Champion' version. I know I would :)
TopicThe Rybka Lounge / Computer Chess / Deep Sjeng 3 update
Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill
Benchmark Name
631.deepsjeng_s
Benchmark Authors
Gian-Carlo Pascutto <gcp [at] sjeng.org>
Benchmark Program General Category
Artificial Intelligence (alpha-beta tree search & patternrecognition)
Benchmark Description
631.deepsjeng_s is based on Deep Sjeng WC2008, the 2008 World Computer Speed-Chess Champion.Deep Sjeng is a rewrite of the older Sjeng-Free program, focused on obtainingthe highest possible playing strength.
It attempts to find the best move via a combination of alpha-beta treesearching, advanced move ordering, ositional evaluation and heuristic forward pruning. Practically, it will explore the tree of variations resulting from a given position to a given base depth, extending interesting variationsbut discarding doubtful or irrelevant ones. From this tree theoptimal line of play for both players ('principal variation') isdetermined, as well as a score reflecting the balance of powerbetween the two.
Specific advances since the previous version include the use of bitboardsbased on Kindergarten/Magic attack calculators, an evaluation functionwith much deeper positional understanding, and a search function that issignificantly better at discarding irrelevant subtrees. The overall programstructure has been optimized for speed as well, dropping support for everythingbesides classical chess.
The rate version of the benchmark is compiled with SMALL_MEMORY.It was observed to require around 700 MiB of memory during pre-release testingon one of SPEC's Linux systems.
The speed version is compiled with BIG_MEMORY, and uses around 10xas much memory as the rate version. Having more memory allows moreinformation to be stored about every position, notably if there are threats,or if there is only one good move. This information then causes some positionsto be investigated deeper, which means it will take longer to reach the samesearch depth, but the result will be more accurate.
Input Description
Both the rate and speed versions use the same inputs, a textfile containingalternations of:
- a chess position in the standard Forsyth-Edwards Notation(FEN) with optional Extended Position Description (EPD) tags, and
- the depth to which this position should be analyzed, inhalf-moves (ply depth).
Output Description
The output consists, per position, of some side information (textualdisplay of the chessboard, phase of the game, used parameters...) followed bythe output from the tree searching module as it progresses. This is formattedas follows:
- Attained depth in half-moves (plies)
- Score for the player that is to move, in equivalents of 1/100thspawn
- Number of positions investigated
- The optimal line of play ('principle variation').
The output is validated against a SPEC-supplied set of expected outputs.Note that the expected output differs between the rate and speed versions.
Programming Language
C++
Known portability issues
None.
Sources and Licensing
Deep Sjeng (which became 631.deepsjeng_s) is licensed directly to SPEC by the author, Gian-Carlo Pascutto.
References
- Sjeng 11.2 & Deep Sjeng: https://www.sjeng.org
- Portable Game Notation Specification (including FEN/EPD):http://www.tim-mann.org/Standard
- Kindergarten/Magic bitboards: Chess Programming Wiki article
Copyright © 2017-2019 Standard Performance Evaluation Corporation (SPEC®)