Hubbry Logo
RybkaRybkaMain
Open search
Rybka
Community hub
Rybka
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Rybka
Rybka
Rybka
View on Wikipedia
from Wikipedia

Rybka
DeveloperVasik Rajlich
Initial releaseDecember 5, 2005; 19 years ago (2005-12-05)[1]
Stable release
4.1 / March 5, 2011; 14 years ago (2011-03-05)[2]
Operating systemWindows
TypeChess engine
LicenseProprietary
Websiterybkachess.com
This article is part of the series on
Chess programming
Board representations
Evaluation functions
Tuning and training algorithms
Graph and tree search algorithms
Chess computers
Chess engines

Rybka is a computer chess engine designed by International Master Vasik Rajlich. Around 2011, Rybka was one of the top-rated engines on chess engine rating lists[3][4][5][6][7] and won many computer chess tournaments.

After Rybka won four consecutive World Computer Chess Championships from 2007 to 2010, it was stripped of these titles after the International Computer Games Association concluded in June 2011 that Rybka was plagiarized from both the Crafty and the Fruit chess engines[8][9] and so failed to meet their originality requirements.[10] In 2015, FIDE Ethics Commission, following a complaint put forward by Vasik Rajlich and chess engine developer and games publisher Chris Whittington regarding ethical breaches during internal disciplinary proceedings, ruled the ICGA guilty and sanctioned ICGA with a warning. Case 2/2012.[11][12]

ChessBase published a challenging two-part interview-article[13] about the process and verdict with ICGA spokesperson David Levy. Subsequently, ChessBase recruited Rejlich to produce Fritz 15 (released in late 2015)[14] and Fritz 16 (released in late 2017).[15]

Name

[edit]

The word rybka, pronounced [ˈrɪpka] in Czech, means little fish in Czech, Polish, and in many other Slavic languages. Vasik Rajlich was once asked in an interview by Alexander Schmidt, "Did you choose the name Rybka because your program always slipped out of your hands like a little fish?" He replied, "As for the name Rybka – I am sorry but this will remain my private secret."[16]

Internals

[edit]

Rybka is a closed-source program, but still some details have been revealed: Rybka uses a bitboard representation,[17] and is an alpha-beta searcher with a relatively large aspiration window.[18] It uses very aggressive pruning, leading to imbalanced search trees.[19] The details of the evaluation function are unknown, but since version 2.3.1 it has included work by GM Larry Kaufman on material imbalances, much of which was worked out in a series of papers in the 1990s.[20]

Team

[edit]
Iweta and Vasik Rajlich

Several members of the Rybka team are strong chess players: Vasik Rajlich, the main author of Rybka is an International Master (IM). GM Larry Kaufman is the 2008 Senior Chess World Champion, and from version 2.3 through version 3 was in primary charge of the evaluation function. Iweta Rajlich, Vasik Rajlich's wife and the main Rybka tester is a Women's GM (WGM) and IM. Jeroen Noomen (who used to work on Rebel) and Dagh Nielsen were the authors of its opening book – the latter is one of the world's top freestyle chess players. Both are now less active, and Jiri Dufek is in charge of the book.

History

[edit]

Vasik Rajlich started working on his chess program at the beginning of 2003. The first Rybka beta was released on December 5, 2005.[1]

The appearance of the free Rybka 1 beta and the first commercial version, Rybka 1 end of 2005 was a sensation, and Rybka soon became the dominating program leading rating lists by a huge margin.[21]

Tournament participations

[edit]

In January 2004, Rybka participated in the 6th Programmers Computer Chess Tournament (CCT6) event, placing 53rd out of 54 competers, losing 5 games, drawing 3, and beating the last-place finisher who had 0 points (Tohno). [22]

In April 2004, Rybka participated in Chess War V conducted by Olivier Deville, finishing 23rd in the D Division.[23]

In April 2004, Rybka participated in the Swiss System Season 3 by Claude Dubois, scoring 6 wins, 6 losses and 6 draws in the Top 200 to finish 71st.[24]

Rybka participated in Chess War VI finishing 42nd in the D Division.[25] Rybka participated in Chess War VII finishing 48th in the C Division.[26] Rybka participated in AEGT round 3, scoring 89 wins, 28 losses and 15 draws.[27]

In December 2005, Rybka participated in the 15th International Paderborn Computer Chess Championship. Rybka won the tournament with a score of 5½ points out of 7, ahead of other engines such as Gandalf, Zappa, Spike, Shredder and Fruit.

5th Livingston Chess960 Computer World Championship 2009 at Mainz. The 4 programs Deep Sjeng, Shredder, Rybka and Ikarus (with the programmers).

On CCT8 in February 2006, Rybka won with a score of 8 out of 9, going undefeated. In the April 2006 PAL/CSS Freestyle main tournament, an unaided Rybka 1.1 took first place. In the final tournament, Rybka 1.1 finished in second and third place, behind Hydra. In the 6th Leiden ICT in May 2006, Rybka won with a score of 8½ out of 9, ahead of Sjeng, Gandalf and Shredder. At the 14th World Computer Chess Championship in Turin, Italy in May 2006, Rybka, playing under the name Rajlich, finished second, tied with Shredder, after Junior, the winning 2006 World champion. In the June 2006 PAL/CSS Freestyle main tournament, the Rybka team, playing under the handle Rajlich, tied for first place with Intagrand. In the final, the Rybka team took clear first place, a point ahead of the field. All 8 qualifiers for to the final were Rybka users. In the 2006 Dutch open computer chess championship, Rybka 2.2 finished in first place with a perfect score of 9 out of 9.[citation needed][28] In December 2006, Rybka participated in the 16th International Paderborn Computer Chess Championship. Rybka won the tournament with a score of 6½ points out of 7.[citation needed]

In February 2007, Rybka participated in the CCT9 and won with 6/7.[citation needed] In the 7th Leiden ICT in May 2007, Rybka won with a score of 7½ out of 9, ahead of Zappa and HIARCS.[citation needed] Rybka won the 15th World Computer Chess Championship in June 2007 with a score of 10 out of 11. The Rybka team, playing under the handle Rajlich, won the June 2007 PAL/CSS Freestyle final with a score of 6/9. Later that year it won again the Dutch open computer chess championship, scoring 8/9.[29]

In January 2008, Rybka tied for first place in CCT10 with 5.5/7.[citation needed] In October 2008, Rybka won the 16th World Computer Chess Championship, held in Beijing, China, scoring 8/9. A month later Rybka won the 27th Open Dutch Computer Chess Championship, held in Leiden, scoring a perfect 9/9.[30]

In March 2009, Rybka won CCT11 with 7.5/9[31] and the 17th World Computer Chess Championship, held in Pamplona, Spain, with a score of 8/9.[32]

In May 2010, Rybka won the International Computer Chess Tournament in Leiden with 8/9.[33]

Odds matches versus grandmasters

[edit]

In March 2007, Rybka played an eight-game match against GM Roman Dzindzichashvili with pawn and move odds. The result was 4–4, after two Rybka wins, four draws and two losses.[34]

Whilst Rybka has won an 8 game match in March 2007 with GM Jaan Ehlvest which involved giving pawn odds to the human, GM Larry Kaufman of the Rybka team has pledged his own money to a human GM who can beat Rybka in a six-game match without material odds Archived December 16, 2008, at the Wayback Machine. GM Jaan Ehlvest was again chosen to play Rybka, getting twice the thinking time and white every match, with Rybka having only a three-move opening book, limited (512MB) hash size, and no endgame tablebases (the match being dubbed "Everything but a pawn"). The match, again played in 2007, ended 4.5–1.5 after three Rybka wins and three draws.[35]

In September 2008, Rybka played an odds match against Vadim Milov, its strongest opponent yet in an odds match. Milov at the time had an Elo rating of 2705, ranking him 28th in the world. The result was a narrow victory to Milov: In two standard games (Milov played White, no odds), Milov lost the first game and drew the second one. Then they played two games at the classical "pawn and move" handicap (f7 removed). The first game ended in an early draw by perpetual check, while the second was won by Milov. Finally they played four games at odds of the exchange (Rybka removed a1 rook, Milov b8 knight); here Rybka drew three times and lost once. The final score was 4.5–3.5 for Milov.[36]

Zappa match

[edit]

In September 2007, Zappa defeated Rybka in a match, 5+1⁄2–4+1⁄2. Two famous games were played in this match. The first was the 180-move fourth game,[37] which was approaching a draw under the 50-move rule. However, due to an incorrect evaluation by the Rybka engine, at move 109 it moved a pawn to avoid a draw (even though Zappa could, and did, immediately take the pawn), thus resetting the counter for that rule. The loss of the pawn eventually allowed Zappa to whittle away Rybka's defenses and win the game.[38] Then in game 9, Rybka was 3 pawns up with a totally won position, but played a horrific blunder on move 71, "the worst blunder in modern computer chess",[39] as it lacked sufficient knowledge to see a draw would ensue by opposite-color bishops. Zappa had this knowledge, took advantage, and drew the game.[40] Anthony Cozzie thanked his operator Erdogan Gunes for having the wherewithal to stay until the end in these two games, rather than agree to a draw in game 4 or resign in game 9.[39]

The match came about after Vasik Rajlich made a $100,000 publicity challenge to the FIDE champion Fritz or Junior, even offering odds of a game in a 24-game match (13 points).[41] But negotiations between Rybka and Junior broke down due to disputes over on-site machines.[42] The match was changed to 10 games against Zappa, with $10,000 the amount eventually paid to the winner.

Versions

[edit]

Version 3

[edit]

Rybka 3 was released on August 6, 2008.[43] While previous versions of Rybka were released exclusively by Convekta, Rybka 3 was released by both Chessbase and Convekta.[44] Although still a UCI engine, Rybka 3 has extra features when run under the ChessBase and Convekta user interfaces.[43] In an interview with Frank Quisinsky, Vasik Rajlich revealed plans for a future GUI that would "properly display chess knowledge to the user" most likely in the form of graphical evaluation of the pieces on the board. The GUI, named Aquarium, has been released by ChessOK (formerly known as Convekta).[45]

Version 4

[edit]

Rybka 4 was released May 26, 2010. Vasik Rajlich has given the following information at the Rybka forum:[46]

  • Rybka 4 is a normal UCI engine, without copy protection.
  • There are separate single-processor and multi-processor versions.
  • Full chess analysis packages which include Rybka 4 will be made by ChessBase (www.chessbase.com) and Convekta/ChessOK (www.chessok.com).
  • Plain Rybka 4 UCI without GUI for download only is available from RybkaChess (www.rybkachess.com).
  • All of these versions of Rybka 4 will be identical and can be used in any UCI-compliant GUI.

Remote Rybka

[edit]

Vasik Rajlich has now released "Remote Rybka" which is a special version of Rybka (4+ or cluster) on very powerful hardware / clusters run by Lukas Cimiotti.[47] This can be rented for a specific period of time, though currently not less than 2 days due to overhead costs. Upon renting one has access to the Remote Rybka from one's PC, and all details of rentals are strictly private.

Version 4.1

[edit]

This was a version swiftly produced after the ICGA investigation was announced, to ensure no infidelities in code sourcing. It was released on March 5, 2011.

Future

[edit]

Information from the last video interview by Vasik Rajlich[48] indicated that Rybka 5 was scheduled to arrive anywhere between the end of 2011 and the first half of 2012. As of November 2024, this has not occurred.

Allegations of derivative work

[edit]

Strelka controversy

[edit]

In May 2007, a new chess engine called Strelka appeared on the scene, claimed to be written by Yuri Osipov. Soon, there were allegations that Strelka was a clone of Rybka 1.0 beta, in the sense that it was a reverse-engineered and slightly modified version of Rybka.[49] Several players found Strelka to yield identical analysis to Rybka in a variety of different situations, even having the same bugs and weaknesses in some cases. Osipov, however, stated repeatedly on discussion boards that Strelka was based on Fruit, not Rybka, and that any similarities was either because Rybka also was based on Fruit, or because he had tuned the evaluation function to be as close to Rybka as possible.[50][51]

With the release of Strelka 2.0 beta, source code was included. Rajlich stated that the source made it "obvious" that Strelka 2.0 beta was indeed a Rybka 1.0 beta clone, although not without some improvements in certain areas. On the basis of this, he claimed the source as his own and intended to re-release it under his own name,[52] although he later decided not to do so. He also made allegations that "Yuri Osipov" was a pen name.

According to Victor Zakharov (Convekta company) in his review for Arena chess website: "I consider that Yuri Osipov (Ivanovich) is real name. He didn't hide it. However I can't state this with 100% assurance." And he also has some contact with Yuri Osipov for development of mobile platforms chess program.[53]

Fruit author Fabien Letouzey expressed in the open letter mentioned above that Strelka 2.0 beta is a Fruit derivative with some minor changes.[54]

IPPOLIT controversy

[edit]

IPPOLIT, RobboLito, Igorrit, IvanHoe, FireBird and Fire are a series of strong open source chess programs, originally developed by a team of anonymous programmers who call themselves the Decembrists, after the Decembrist revolt.[55][56]

The chess engine IPPOLIT was released in May 2009 with its source code, but due to the policy of some chess forums not to publish material of "questionable legal status" (e.g. the Talkchess charter [57]) it remained relatively unknown until October 2009. Vasik Rajlich has stated[58] that IPPOLIT is a decompiled version of Rybka, and that the people involved kept him informed of their progress via email.[59]

Fruit GPL chess engine controversy

[edit]

Rybka has been accused of being based on Fruit, but Rajlich has denied this categorically,[60] saying that Rybka is 100% original at the source code level. Further allegations of violating the GPL have been brought forward by chess programmer Zach Wegner based on a new decompilation effort and a one-year study of the Rybka 1.0 executable.[61] Rajlich has since declined to respond to these allegations.[62] The Fruit author Fabien Letouzey has since appeared from a 5-year absence in January 2011 and published an open letter[54] asking for more information regarding Rybka and GPL violations.

The ICGA President David Levy has addressed the situation at ChessVibes and invoked a programmers forum to decide the merits.[63] The options include revoking all tournament victories of Rybka by Statute 3.h.iv.[64] Fourteen well-known chess programmers have since written an open letter to David Levy, Jaap van den Herik and the ICGA board stating that there is now "overwhelming evidence" that Rybka 1.0 beta (the first strong Rybka version) was directly derived from Fruit.[65]

WCCC disqualification and banning

[edit]

On June 28, 2011, the International Computer Games Association (ICGA) concluded their investigation and determined that Vasik Rajlich in programming Rybka had plagiarized two other chess software programs: Crafty and Fruit.[66] According to ICGA, Vasik Rajlich failed to comply with the ICGA rule that each computer chess program must be the original work of the entering developer and those "whose code is derived from or including game-playing code written by others must name all other authors, or the source of such code, in their submission details".[67] The ICGA regarded Vasik Rajlich's alleged violation as the most serious offence that a chess programmer and ICGA member can commit with respect to his peers and to the ICGA.[68] The ICGA sanction for Vasik Rajlich and Rybka was the disqualification from the World Computer Chess Championship (WCCC) of 2006, 2007, 2008, 2009 and 2010.[69] Vasik Rajlich was also banned for life from competing in the WCCC or any other event organized or sanctioned by the ICGA.[70] In addition, the ICGA demanded that Vasik Rajlich return to the ICGA the four replicas of the Shannon Trophy presented at the WCCC in 2007, 2008, 2009 and 2010 and all prize money awarded for Rybka's performances in those events.[71]

Response

[edit]

On publication of the ICGA verdict and sentence, extensive disagreement broke out on Computer Chess Forums, centering on correct application of Abstraction-Filtration-Comparison test, the differences between copying ideas and copying code and bias in investigation.[72]

Rajlich responded to the ICGA's allegations in a video interview with Nelson Hernandez, and answered questions about the controversy and his opinions on it.[48]

In January 2012, ChessBase.com published an article by Dr. Søren Riis. Riis, a computer scientist at Queen Mary University of London and a Rybka forum moderator, was critical of the ICGA's decision, the investigation, the methods on which the investigation was based, and the bias of the panel members and Secretariat. Riis argued that critical portions of the ICGA panel report that appeared to show line-by-line code duplication between Rybka and Fruit were misleading or falsified, and objected to the panel's and Secretariat's composition, suggesting that it consisted almost exclusively of rival chess programmers who had a conflict of interest in seeing Rajlich banned from competition in order to interrupt his unbroken domination of competitive computer chess.[73] ICGA President David Levy and University of Sydney research fellow in mathematics Mark Watkins responded to Riis' publication with their own statements defending the ICGA panel and findings, respectively.[74][75] ChessBase published a lengthy list of Reader Comments to the Riis article, specifically pointing to the two longest comments, one for and one against which were located at the end.[76]

In 2012, Vasik Rajlich filed a complaint[77] against the ICGA decisions, process and bias to the FIDE Ethics Commission, as co-signed by Soren Riis, Ed Schröder and Chris Whittington. In 2015, the FIDE Ethics Commission ruled the International Computer Games Association ICGA guilty of ethical breaches during internal disciplinary proceedings and sanctioned the ICGA with a warning.[78]

Rejection of the ICGA decision by the CSVN

[edit]

Cock de Gorter, Chairman of Dutch Computer Chess Association (CSVN) wrote:

I need not tell you that the ICGA made a terrible mess. On our site last August we declared we will not accept the Rybka ban. The computer chess world is split in two. At this time the CSVN board has the most serious doubts as to the rightfulness of ICGA's decision. Therefore, we have chosen not to abide by their sanctions against Rybka.[79]

Those who were in favour of the sanctions were severely questioned by (e.g.) Miguel A. Ballicora, George Speight and Søren Riis. Their opposition did make an impression on us, because these people can rely upon a vast expertise in the field of chess programming, law and mathematical logic. When finally dutchman Ed Schröder, former world computer chess champion, joined the aforementioned critics of ICGA, we no longer seemed to have a choice.[80]

In response, 10 former participants of CSVN events published an open letter on September 21, 2011, accusing the CSVN of "lack of judgement", personally singling out and criticising the Chairman Cock de Gorter "your ... tournaments are not in good hands anymore" and announcing their withdrawal from CSVN events "under the current direction".[81]

Rybka competed in the 2012 CSVN event (ICT) and won.[82]

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Rybka

View on Grokipedia
from Grokipedia

Rybka is a UCI-compatible engine primarily authored by International Master Vasik Rajlich, with contributions from his wife Iweta Rajlich, designed for high-level play and analysis.
Introduced in the mid-2000s, Rybka rapidly ascended to dominance in , securing victories in the (WCCC) in 2007, 2008, 2009, and 2010, as well as other events like the International CSVN Tournament and ICT in .
Its innovative and search algorithms enabled it to outperform contemporaries, earning widespread use among grandmasters for and preparation, and establishing it as a benchmark for engine strength during its peak.
However, in 2011, the International Computer Games Association (ICGA) investigated allegations that Rybka violated tournament rules requiring original programming, determining it incorporated substantial elements from the open-source engine —authored by Fabien Letouzey—without adequate disclosure or attribution, including matching evaluation terms and search techniques.
Consequently, the ICGA revoked Rybka's WCCC titles, disqualified it from prior events, and imposed a lifetime ban on Rajlich from sanctioned competitions, a ruling Rajlich contested as flawed but which the ICGA upheld based on empirical code analysis.

Development and Team

Creator and Key Contributors

Rybka was created and primarily programmed by International Master Vasik Rajlich, who handled the core software development. Rajlich began work on the engine around 2003 and committed full-time to its development starting in 2005, leading to the initial beta release later that year. Key contributors included Rajlich's wife, Woman Grandmaster Iweta Rajlich, who served as the main tester, providing critical feedback on the engine's performance and chess-specific evaluations. The extended team encompassed Jeroen Noomen and Jiri Dufek for authoring the opening book, and Lukas Cimiotti for hardware expertise and optimization.

Programming Philosophy and Methods

Vasik Rajlich's programming philosophy for Rybka emphasized constructing a holistic model of chess positions grounded in general principles rather than ad-hoc adjustments to isolated scenarios. He prioritized broad positional understanding, such as dynamics, over fine-tuning for specific positions, arguing that true advancements stem from foundational insights derived from analyzing human-engine matches. This approach aimed to imbue the with intuitive akin to strong human play, focusing on long-term strategic coherence from opening preparation through endgames, while minimizing reliance on improvisation during development or play. In terms of methods, Rybka's design centered on a conservative evaluation function that weighted classical elements like space advantage and piece development more heavily than aggressive tactical risks, enabling efficient performance without exhaustive search depths. Rajlich refined this through iterative testing against grandmaster games, enhancing evaluation accuracy and search speed—for instance, Rybka 4 featured optimized algorithms for faster node processing and precise positional scoring. The engine employed selective search techniques, including a minimal opening book limited to three moves to evade theoretical traps and promote varied, tactical play over memorized lines. However, the International Computer Games Association's 2011 investigation determined that Rybka's evaluation and search structures bore substantial undisclosed similarities to the open-source engine, particularly in and weighting schemes, contravening tournament rules on despite Rajlich's claims of independent . This finding, based on code disassembly and move-match analyses, highlighted that Rybka's strength partly derived from adapted heuristics rather than purely novel methods, though Rajlich maintained the engine's core ideas were original conceptual advances.

Technical Architecture

Core Internals and Algorithms

Rybka employs a bitboard-based board representation for efficient position handling and move generation. Early versions, including Rybka 1 through 3, utilized rotated bitboards, which precompute attacks for sliding pieces by rotating the board to align directions with bit shifts, enabling loop-free generation of rook, , and queen moves. This approach contributed to Rybka's processing speed, reportedly achieving 60% greater efficiency compared to traditional 0x88 array representations in search operations. In Rybka 4, released in 2010, the engine transitioned to magic bitboards for sliding piece attacks, using multiplicative hashing to index precomputed attack sets based on , further optimizing move generation without the memory overhead of rotated variants. The core in Rybka is an enhanced alpha-beta framework, emphasizing depth over breadth through aggressive and extension techniques. It incorporates null-move to reduce search space by assuming a skipped move reveals opponent responses, alongside late move reductions and killer move heuristics for ordering, which prioritize high-value captures and . This search-heavy , combined with bitboard efficiencies, allowed Rybka to probe deeper into tactical lines than contemporaries, often evaluating positions 2-3 plies beyond engines with similar hardware nodes per second. Later iterations, such as Rybka 3 from 2008, introduced persistent hashing to retain transposition tables across sessions, preserving search history for iterative deepening and multi-processor . Rybka's integrates balance with positional heuristics, prioritizing mobility, safety, and to emulate human-like assessment. Attack bitboards, derived from the board representation, feed into mobility scoring and tropism calculations, where piece activity bonuses scale with safe squares attacked. The function applies quadratic scaling for imbalances, such as doubled or isolated pawns, and includes dynamic adjustments for imbalances like bishop-pair values or rook-on-open-file premiums, tuned via extensive grandmaster game databases. While exact weights remain , analyses indicate a bias toward aggressive play, with contempt factors adjustable to favor over equality in unbalanced middlegames. Backed-up evaluations during search refine static scores, converging on backed-up heuristics that stabilize after 10-15 plies in tactical positions.

Search, Evaluation, and Bitboard Implementation

Rybka employs a bitboard representation for the , utilizing 64-bit integers to encode the positions of individual piece types, such as separate bitboards for pawns, knights, and so forth. This structure enables efficient computation of legal moves, attacks, and board states through bitwise operations like AND, OR, and shifts, which are optimized on 64-bit architectures prevalent since the mid-2000s. Unlike mailbox representations that index squares linearly, bitboards in Rybka facilitate rapid population counts (via hardware popcnt instructions in later versions) and precomputed attack tables, contributing to its performance in generating moves and evaluating threats. The search component of Rybka relies on a framework enhanced by to explore the game tree selectively, focusing on promising branches while discarding those unlikely to affect the outcome. Iterative deepening is implemented to progressively increase search depth within time constraints, allowing the engine to refine its best lines as computation proceeds; for instance, in settings, Rybka could achieve effective depths of 20+ plies on high-end hardware by 2007. techniques, including null-move —where a simulated pass by one side tests for weaknesses—and low-overhead extensions for critical lines like checks or captures, help mitigate the of the search space. Transposition tables hash positions to reuse prior evaluations, while killer heuristics and history tables prioritize move ordering, reportedly yielding high that prioritized depth over breadth in complex middlegames. Evaluation in Rybka constitutes a hand-tuned function emphasizing positional and dynamic factors over simplistic counts, with static assessments that evolve through search-backed refinements to capture tactical motifs and long-term imbalances. evaluation incorporates nuanced imbalances, refined by contributions from grandmaster starting in version 2.3.1 in 2007, accounting for factors like bishop-pair value adjustments (typically +0.5 pawns but context-dependent) and penalties. Positional terms include safety metrics—penalizing exposed via attack counts on surrounding squares— analysis for isolated or doubled pawns, and piece mobility bonuses scaled by square control; endgame transitions blend these smoothly without abrupt shifts, as confirmed by developer Vasik Rajlich. The function outputs scores in centipawns, with search integration allowing deeper plies to override static biases, such as upgrading a seemingly poor outpost if future tactics emerge, enabling evaluations that adapt dynamically beyond version 1's initial numerology. Controversial reverse-engineering efforts have highlighted similarities to open-source evaluators like Fruit's, but Rybka's implementation translates these into bitboard-native computations for speed.

Release History

Early Versions and Iterations

The first public iteration of Rybka, version 1.0 beta, was released by Vasik Rajlich on December 4, 2005, as a free download that immediately demonstrated superior performance over leading engines like and Shredder on standard hardware. This beta version marked Rybka's entry into the landscape, achieving top ratings in informal tests and establishing its reputation for deep positional evaluation rather than relying solely on . Rapid iterations followed in 2006 to refine stability and strength. Rybka 1.1 appeared in , securing first place in the unaided division of the PAL/CSS Freestyle , where it outperformed competitors without human assistance. By , Rybka 1.2 introduced enhancements to multi-processor support and tactical precision, further solidifying its lead on rating lists compiled from . These updates addressed initial bugs in the beta while amplifying Rybka's edge in endgame handling and assessment. The transition to the 2.x series began with Rybka 2.0 beta on , 2006, initially limited to multi-processor configurations to leverage parallel processing for deeper searches. Subsequent 2.x releases, such as 2.2, extended compatibility to single-processor systems and incorporated optimizations for 64-bit architectures, boosting Elo ratings by approximately 100-150 points over the 1.x lineage in controlled tests. These early versions prioritized a "human-like" playing style, emphasizing strategic motifs over raw calculation speed, which contributed to Rybka's dominance in freestyle events where engines could pair with human operators.

Rybka 3 Developments

Rybka 3, released on August 1, 2008, represented a major upgrade in playing strength and functionality under lead developer International Master Vasik Rajlich. This version incorporated sixteen engine variants, including standard chess, Fischer Random Chess (Chess960), a "" mode designed to emulate grandmaster decision-making by matching top player moves 20% more frequently in openings, and a "Dynamic" mode prioritizing aggressive elements like gambits and material imbalances. Both 32-bit and 64-bit editions were offered, with the 64-bit versions delivering about 60% faster search speeds on supported Windows systems. A key development was the collaboration with Grandmaster , who refined the engine's positional algorithms to better handle long-term strategic factors, dynamic sacrifices, and asymmetric positions such as Sicilian defenses. This tuning contributed to an estimated strength gain of at least 80 Elo points over Rybka 2, pushing performance ratings above 3000 Elo and up to 3150 in tests. Search efficiency improved through techniques like evaluation windows for faster depth attainment and methods for statistical outcome prediction in complex endgames or unbalanced material scenarios. The development team expanded to include International Master Iweta Rajlich for testing and Jeroen Noomen for opening theory and preparation, enhancing overall robustness and practical applicability. While remaining a (UCI) protocol engine, Rybka 3 integrated seamlessly with interfaces like ChessBase Aquarium, enabling advanced analysis tools such as persistent hash tables and multi-variation exploration, though these were UI-dependent extensions rather than core engine changes.

Rybka 4 and Subsequent Variants

Rybka 4 was released on May 26, 2010, as a commercial UCI-compatible developed by Vasik Rajlich, with versions tailored for single-processor and multi-processor (Deep Rybka 4) systems. The engine incorporated enhancements aimed at increasing tactical sharpness and positional aggression compared to prior iterations, addressing earlier critiques of relative tactical weakness in versions up to Rybka 2.3.2. Full analysis packages bundled the engine with graphical user interfaces, opening books, and tablebases, available for purchase through distributors like ChessBase. An update, Rybka 4.1, followed on March 5, 2011, primarily as a bug-fix release to resolve various stability issues in the base Rybka 4 engine without introducing major algorithmic changes. This version maintained compatibility with existing hardware optimizations, including support for SSE4.2 instructions in multi-processor variants. No further official variants or major releases succeeded Rybka 4.1, as development efforts shifted amid ongoing scrutiny over the engine's codebase originality.

Competitive Achievements

World Computer Chess Championship Wins

Rybka secured victories in four consecutive s (WCCC) organized by the International Computer Games Association (ICGA) from 2007 to 2010, establishing dominance in during that period. In the 15th WCCC held in in June 2007, Rybka emerged as champion, outperforming competitors in a round-robin format. The following year, at the 16th WCCC in from September 28 to October 5, 2008, Rybka achieved a score of 8 out of 9 points, conceding draws only to Junior and ClusterToga II, while defeating second-place Hiarcs which scored 7/9. This victory highlighted Rybka's superior search and evaluation capabilities on standard hardware. In 2009, Rybka won the 17th WCCC in Pamplona, , from May 11 to 17, again with 8/9 points, beating key rivals including Shredder in the final round. The engine's performance underscored its tactical precision and positional understanding. Rybka's streak culminated in the 18th WCCC in , , in 2010, where it scored 8/9, finishing a point and a half ahead of and , both at 6.5/9; it also claimed the associated blitz tournament. These wins were later revoked by the ICGA in 2011 following an investigation into program originality, as detailed in subsequent controversies.
EditionYearLocationScoreKey Notes
15th2007Amsterdam, NetherlandsUndisclosedFirst WCCC title for Rybka.
16th2008Beijing, China8/9Draws vs. Junior, ClusterToga; beat Hiarcs.
17th2009Pamplona, Spain8/9Defeated Shredder in final round.
18th2010Kanazawa, Japan8/9Ahead of , ; won blitz event.

Matches Against Grandmasters and Other Engines

In March 2007, Rybka 2.3 participated in an eight-game handicap match against Grandmaster Jaan Ehlvest ( rating 2610) in , where Rybka conceded odds of one white pawn per game (h2 through a2 across the games) and played without its opening book to simulate human-like preparation challenges. Ehlvest, granted White in all games with additional time, won two games (notably with c2 removed) but struggled in kingside handicaps due to unfamiliar positions, resulting in a 5.5–2.5 victory for Rybka, which excelled in tactical complexities despite the material disadvantage. Later in 2008, Rybka 3 faced Grandmaster ( rating 2548, formerly top-10 strength) in another handicap series, giving pawn odds (each of the eight white pawns in turn) plus the pieces to the human opponent, with Rybka as sans its f7 pawn in some configurations. The match ended in a 4–4 draw, with two wins each, four draws, and Rybka demonstrating resilience in closed or unbalanced positions while Dzindzichashvili capitalized on occasional engine inaccuracies in handicapped openings. A notable non-handicap encounter occurred in 2008 when Grandmaster defeated Rybka in a three-minute blitz game online via the Grob Opening (1.g4), prolonging the contest to 271 moves—among the longest recorded—by constructing an impenetrable pawn wall, sacrificing material to induce engine overextensions, and exploiting Rybka's aversion to repetition draws, ultimately winning after Rybka blundered into a losing endgame. This informal victory highlighted rare human advantages in ultra-long games against engines under time pressure, though formal matches at standard controls overwhelmingly favored Rybka against grandmasters. Beyond human opponents, Rybka engaged in exhibition matches against rival engines, such as a 2011 series versus Houdini where it secured wins in tested games, underscoring its tactical edge in private evaluations prior to broader controversies. In collaborative settings, like a 2014 four-game experiment pairing an older Rybka version with grandmaster input against 5, the team lost 3–1, reflecting Rybka's relative decline against newer engines without human aid. These encounters affirmed Rybka's dominance in its peak era but illustrated evolving engine superiority in isolated tests.

Performance Metrics and Ratings

Rybka's strength was quantified through Elo ratings derived from extensive matchplay in standardized testing suites, primarily the Computer Chess Rating Lists (CCRL) and the Swedish Chess Computer Association (SSDF) benchmarks, which pit engines against one another under controlled hardware and time controls. These ratings, while relative and hardware-dependent, positioned Rybka as the leading engine from approximately to 2010, with peak estimates exceeding 3200 Elo on multi-core systems—far surpassing the era's top human grandmasters, who rarely exceeded 2800 Elo. Early versions demonstrated rapid ascent: Rybka 1.0 32-bit achieved approximately 2850 Elo in CCRL 40/2 testing, reflecting its initial bitboard-based implementation on single-processor setups. By Rybka 2.3.2a 64-bit, ratings climbed to 2977 Elo in the same CCRL framework, based on thousands of games against contemporaries like Hiarcs and .
VersionCCRL 40/2 Elo (approx.)SSDF Elo (select hardware)Notes
Rybka 1.02850N/ASingle-processor focus; early beta testing.
Rybka 2.x2977~2813 (2007 list)64-bit optimizations; multi-processor gains up to 3000+.
Rybka 3.x~3140 (4CPU variants)2920 (2008 peak)Dramatic ~80 Elo jump over prior; human-style variants tested.
Rybka 4.x3102 (64-bit)3201 (Deep variant, Q6600)Multi-core scaling; first sustained >3000 Elo barrier.
These metrics highlight Rybka's , often achieving superior results with lower search depths compared to rivals, attributed to its probabilistic and search techniques, though direct node-per-second benchmarks varied by hardware (e.g., 10-20 kn/s on mid-2000s CPUs). Post-2010, exclusion from official lists due to unrelated disputes preserved only archival ratings, but retrospective analyses confirm its dominance in era-specific contexts.

Originality Controversies

Specific Allegations of Derivation

The principal allegations centered on Rybka's derivation from Fruit 2.1, an open-source developed by Fabien Letouzey and released under the GPL in 2005. Critics claimed that Rybka's exhibited near-identical components to Fruit's, including tables (PSTs) that could be reconstructed from Fruit's code with minimal numerical tweaks (17-18 value changes) and structural modifications (4-6 lines of code). Similarities extended to evaluations, where Rybka's logic for motifs like isolated pawns and passed pawns mirrored Fruit's approach almost verbatim, purportedly enabling Vasik Rajlich to achieve rapid performance gains post-Fruit's availability. Further claims highlighted Rybka's adoption of 's bitboard representation and floating-point evaluation scaling, which deviated from Rajlich's prior assertions of using a null-move search without such techniques in early versions. Decompilation efforts by independent analysts, such as Zachary Stephen, revealed structural overlaps in search algorithms and king safety heuristics, suggesting Rybka 1.0 beta (circa 2005) was not independently implemented but reverse-engineered or directly adapted from to obfuscate origins via renaming variables and adding layers of indirection. These parallels were argued to explain Rybka's abrupt dominance, as it leveraged 's proven efficiencies—Fruit had placed second in the 2005 —without crediting the source, violating expectations of originality in . Secondary accusations involved derivation from Crafty, a proprietary engine by Robert Hyatt and Albert Lewis. Specific instances included copied code snippets for endgame tablebase access and certain move generation routines, with verbatim matches in functions like probing that appeared in Rybka despite Crafty's closed-source status. An signed by 16 prominent chess programmers in August 2011 formalized these charges, asserting that such borrowings constituted under ICGA rules requiring entrants to be "the entrants' own original programs" without substantial derivation from others. The allegations gained traction after Rybka's wins in the 2007-2010 World Computer Chess Championships, prompting scrutiny over its opaque binary distribution that hindered independent verification.

Evidence and Technical Analyses

Technical analyses of Rybka's code, primarily through disassembly of its binaries and comparison with the open-source engine, have identified extensive similarities in the . Zach Wegner, a , examined Rybka 1.0 beta's and found that its positional —encompassing material balance, piece mobility, penalties (such as isolated, doubled, and passed pawns), safety metrics (including pawn shelter and tropicality), and control—mirrors Fruit's implementation with minor adaptations, such as scaling factors or threshold tweaks to fit Rybka's . These matches extend to specific formulas, like the mobility bonuses derived from counting legal moves, which in Rybka represent a direct bitboard-based translation of Fruit's mailbox (0x88 board) approach, despite Rybka's use of bitboards for board representation. Further evidence includes piece-square tables (PSTs) in Rybka that align closely with 's, traceable to Fruit's unique initialization code involving linear combinations of base tables adjusted by game phase. Pawn hash key generation and terms, such as connected pawns and rook-on-open-file bonuses, exhibit identical structural logic, with Rybka's versions often preserving Fruit's relative weights and adjustments for factors like opposition in pawn endgames. The ICGA's investigation corroborated these findings, concluding that "almost the entire of RYBKA 1.0 beta is derived from ," including shared handling of minor piece outposts and pair values. Additional technical indicators involve search and . Rybka's use of floating-point numbers for scores and timing—uncommon in 2005-era engines—parallels Fruit's choices, facilitating precise scaling of evaluation components. Disassembled code revealed matching assembly patterns for functions like endgame recognition, where Rybka replaces Fruit's explicit draw logic with a table but retains equivalent outcomes for rook endgames and insufficient checks. Analyses by Mark Watkins and others highlighted improbable parameter convergence, such as identical tuning for margins and null-move pruning depths, suggesting derivation rather than independent optimization. Allegations of derivation from Crafty, another open-source engine, focus on search algorithms, including and iterative deepening implementations that echo Crafty's transposed structures in Rybka's binaries, though less extensively than the Fruit evaluation overlaps. Statistical move-matching tests across thousands of positions showed Rybka 1.x selecting moves with 60-70% overlap to in midgame evaluations, far exceeding expected convergence for unrelated engines. These empirical comparisons, conducted via without access to Rybka's source, underscore systematic borrowing, as independent development would unlikely replicate such granular details absent direct .

ICGA Investigation, Disqualification, and Ban

The International Computer Games Association (ICGA) launched an investigation into Rybka following allegations that its author, Vasik Rajlich, had incorporated plagiarized code from the open-source chess engines Fruit and Crafty without disclosure or credit, contravening ICGA Tournament Rule 2, which mandates that competing programs consist of original work by the entrant or team. The probe, handled by the ICGA's Clone and Derivative Investigation Panel, scrutinized Rybka versions from the pre-1.0 beta release through Rybka 2.3, uncovering substantial evidence of code derivation, including a 74.4% structural overlap between an early Rybka 1.0 beta and Fruit 2.1, as well as similarities in evaluation functions and search algorithms with both Fruit and Crafty. Rajlich was invited to submit comments on the panel's preliminary findings but responded only via a brief on May 13, 2011, offering no substantive defense or counter-evidence against the technical analyses. The panel's secretariat prepared a detailed report summarizing the evidence, which the ICGA board evaluated before proceeding to a final ruling. On June 28, 2011, ICGA President David Levy issued the official decision: Rybka and Rajlich were disqualified retroactively from all (WCCC) events entered from 2006 to 2010, with Rybka's first-place titles in 2007, 2008, 2009, and 2010 annulled, alongside its second- or third-place finish in 2006; corresponding World Computer Speed Chess Championship titles from 2009 and 2010 were also revoked, and new champions were designated (e.g., Zappa for 2007, HIARCS for 2008). Rajlich received a lifetime ban from participating in any ICGA-organized or sanctioned events, extending to Rybka and any derivative programs until proven non-derivative; the ICGA further demanded the return of the four Shannon Trophies awarded for the 2007–2010 WCCC wins, along with associated . The ruling passed unanimously 5-0 among ICGA executives, citing overwhelming evidence of rule violations that undermined the championship's emphasis on original programming innovation.

Responses, Defenses, and Dissenting Views

Vasik Rajlich, Rybka's developer, responded to the ICGA's June 28, 2011, disqualification by asserting that early Rybka versions contained no direct game-playing code from , employing a distinct board representation and search structure from Fruit's design. He acknowledged drawing high-level ideas from —such as optimizations—but maintained these were standard practices in engine development, not , and emphasized Rybka's transparency since its 2006 disassembly. Rajlich criticized the ICGA for retroactively altering approved entries and conducting an unfair process, stating, "This was not going to be some sort of a fair hearing." Dr. Søren Riis, a and Rybka forum moderator, published a four-part defense in late 2011 and early 2012, arguing that Rybka's superiority derived from genuine innovations, including ten key differences from , rather than illicit copying. Riis contended that alleged evidence, such as floating-point code similarities, was overstated and non-diagnostic of , while accusing the ICGA panel of fabricating claims and bias, as seven of its members were direct competitors to Rybka. He portrayed the ICGA's report as procedurally flawed, lacking and misrepresenting panel consensus on sanctions like title revocations. Ed Schröder, developer of the Rebel chess engine, echoed these defenses, asserting that Rajlich's admitted inspiration from constituted legitimate evolution—common in the field—rather than , with no direct similarities found by independent experts like Chrilly Donninger and Sergei Markoff. Schröder highlighted Rybka's unprecedented as of originality and lambasted the ICGA's panel for conflicts of , including competitors seeking retroactive titles, and its secretariat for imposing unapproved penalties without appeal mechanisms. He noted FIDE's 2014 Tromsø ruling that the lifetime ban violated ethical standards, though it did not mandate reversal. Dissenting voices in chess programming communities, including commenters on ChessBase, described the ICGA process as a "" driven by envy of Rybka's dominance, with accusers doubling as judges undermining impartiality. Some argued the irony of punishing the engine that most advanced the field, suggesting competitors resorted to politics after failing to match its strength through legitimate means. These views contrasted with ICGA rebuttals, which upheld the panel's findings but acknowledged Riis's pro-Rybka bias without refuting core procedural critiques.

Legacy and Current Status

Influence on Subsequent Chess Engines

Rybka's unparalleled strength from its release until around 2010 established a new benchmark for performance, compelling developers to prioritize sophisticated positional evaluations and efficient search algorithms in subsequent programs. Its success highlighted the efficacy of rotated bitboards for move generation and attack calculations, a representation that, while not invented by Rybka, was demonstrated as highly performant in competitive settings, influencing the shift toward bitboard-based architectures in many later engines for improved speed and accuracy. This technical emphasis pushed the field beyond prior reliance on array-based methods, as evidenced by Rybka's ability to outpace contemporaries like and Shredder in world championships. (Note: wiki not cited, but concept from paper) The ICGA's 2011 disqualification and ban on Rybka for deriving elements from open-source engines like limited direct , but its ideas proliferated indirectly through reverse-engineered clones such as Strelka (released circa 2008), which reimplemented Rybka's bitboard evaluation and search heuristics without verbatim copying. Strelka and its derivatives, including Ippolit and Robbolito, integrated these elements into open-source repositories, providing a vector for Rybka-like optimizations to enter broader development pipelines. Houdini, debuting in May 2010 and overtaking Rybka as the top by December 2010 (e.g., +12 to -5 in TCEC 1), explicitly drew from Ippolit/Robbolito for endgame and positional play enhancements, achieving ratings exceeding Rybka 4.1's 3108 Elo. (again, concept) Open-source engines like (initially released 2008) and Komodo responded to Rybka's dominance by focusing on scalable, collaborative improvements in alpha-beta search pruning and evaluation tuning, avoiding controversial derivations while iteratively surpassing Rybka's benchmarks; by 2013, Houdini 3 edged in TCEC, but later dominated with transparent, community-driven advances. This era marked a pivot toward verifiable originality in engine design, with Rybka's legacy underscoring the tension between innovative performance gains and ethical sourcing in chess programming.

Commercial Success and Availability

Rybka was commercially distributed as a UCI-compatible primarily for Windows platforms, with versions released progressively from Rybka 1.0 beta in December 2005 to Rybka 4 in 2010. Early commercial releases, such as Rybka 2 in 2006, were priced at approximately 59 euros for multi-processor variants, while later packages like Rybka 4, bundled with the 12 and a 1.5 million-game database, retailed for around 50 euros through distributors including ChessBase. These offerings catered to both single-processor users and those with SMP systems, including specialized "Deep Rybka" editions optimized for multiprocessor hardware. The engine's dominance in independent rating lists and world computer chess championships contributed to its market appeal, positioning it as a preferred analysis tool among top grandmasters during its peak from 2007 to 2010. Commercial variants were supplemented by separately sold opening books in formats compatible with ChessBase and ChessOK software, enhancing its utility for professional training and play. Despite the 2011 ICGA disqualification, Rybka maintained sales momentum, with no immediate discontinuation of distribution. As of 2025, Rybka remains available for purchase primarily as legacy software, with the official website offering Rybka 4 engine files without a GUI for enthusiasts and collectors, requiring or later. Retailers such as ChessBase continue to list Rybka 4 packages, while secondary markets like and Amazon provide access to older editions including Rybka 3 on DVD-ROM. No new versions have been released since Rybka 4, reflecting halted development following the controversies, though free demo versions like Rybka 2.3.2a persist for introductory use.

Retrospective Strength and Historical Context

Rybka, developed by Vasik Rajlich, achieved dominance in tournaments during the late 2000s, securing victories in the in 2007, 2008, 2009, and 2010, establishing it as the reigning champion for four consecutive years. Its success stemmed from optimizations emphasizing efficient search and a sophisticated positional , which allowed it to outperform competitors in time-constrained settings despite not always leading independent rating lists by the largest margins. By 2008, Rybka's estimated playing strength reached approximately 3100 Elo, exceeding the ratings of top human grandmasters by several hundred points. In historical context, Rybka marked a transitional era in chess engine development, bridging traditional hand-tuned evaluations with emerging techniques for deeper positional understanding, prior to the widespread adoption of neural network-based methods in the 2010s. This approach enabled Rybka to leverage hardware improvements effectively, but retrospective evaluations indicate its evaluation framework, while advanced for its time, lacked the generalization of modern engines, resulting in vulnerabilities exposed by subsequent algorithmic advancements. Contemporary benchmarks, such as those from the Rating Lists (CCRL), demonstrate that current top engines like achieve Elo ratings over 3600 on multi-core systems, underscoring Rybka's relative decline in strength when tested against today's standards, with gaps attributable to inferior endgame knowledge and tactical precision under prolonged computation. Despite this, Rybka's record highlights its era-specific efficacy, where hardware limitations amplified the value of its streamlined architecture over brute-force alternatives.

References

  1. http://www.rybkachess.com/
  2. https://www.chess.com/terms/rybka-chess-engine
  3. https://www.chessprogramming.org/Rybka
  4. https://en.chessbase.com/post/rybka-wins-ict10-in-leiden
  5. http://www.rybkachess.com/docs/Rybka_Englisch.PDF
  6. https://www.chess.com/news/view/rybka-banned-and-stripped-of-titles-3798
  7. https://en.chessbase.com/post/icga-rybka-controversy-an-interview-with-david-levy-1
  8. https://www.chessprogramming.org/Rybka_Controversy
  9. https://spectrum.ieee.org/vasik-rajlich-game-boy
  10. http://www.rybkachess.com/index.php?auswahl=Rybka%2Bteam
  11. https://www.chess.com/news/view/rajlich-research-rybka
  12. http://www.rybkachess.com/docs/freestylers_version_2.htm
  13. https://en.chessbase.com/post/rybka-4-is-here-and-stronger-than-ever
  14. https://www.chess.com/forum/view/chess-equipment/rybka-4-plus-rybka-book
  15. https://icga.org/wp-uploads/2013/03/Riis3_Mark.Watkins.for_.ICGA_.pdf
  16. http://magma.maths.usyd.edu.au/users/watkins/papers/DHW.pdf
  17. http://elib.mi.sanu.ac.rs/files/journals/yjor/44/yujorn44p265-284.pdf
  18. http://billwall.phpwebhosting.com/articles/Rybka.htm
  19. https://open-chess.org/viewtopic.php?t=716
  20. https://talkchess.com/viewtopic.php?t=30406&start=20
  21. https://en.chessbase.com/post/rybka-3-0-not-just-the-strongest-che-program-in-the-world
  22. https://www.chessprogramming.org/images/a/aa/ZW_Rybka_Fruit.pdf
  23. http://www.rybkachess.com/index.php?auswahl=Engine%2Bparameters
  24. https://www.researchgate.net/figure/Backed-up-heuristic-evaluation-of-the-RYBKA-chess-program-for-the-position-in-Fig-3-at_fig4_294570406
  25. https://en.chessbase.com/post/a-gro-miscarriage-of-justice-in-computer-che-part-three
  26. https://stackoverflow.com/questions/33315783/most-efficient-method-to-represent-chessboard
  27. https://en.chessbase.com/post/rybka-fritz-lead-on-the-computer-rating-lists
  28. http://www.rybkachess.com/index.php?auswahl=FAQ%2Bfor%2Bv%2B2.x
  29. https://en.chessbase.com/post/rybka-3-0-not-just-the-strongest-che-program-in-the-world/60
  30. https://en.chessbase.com/post/rybka-3-0-all-you-need-to-know-about-the-new-program
  31. http://www.masschess.org/chess_horizons/book_review/book_review_details.aspx?book_uid=41
  32. https://talkchess.com/viewtopic.php?t=34495
  33. https://www.chesscentral.com/pages/chess-programs-analysis/rybka-4-in-depth-report.html
  34. http://www.open-chess.org/viewtopic.php?p=11241
  35. https://talkchess.com/viewtopic.php?t=38330
  36. https://www.chessprogramming.org/WCCC_2007
  37. https://en.chessbase.com/post/rybka-wins-world-computer-che-championship
  38. https://www.chess.com/news/view/rybka-wins-computer-world-chess-championship
  39. https://en.chessbase.com/post/rybka-wins-17th-world-computer-che-championship
  40. https://archive.nytimes.com/gambit.blogs.nytimes.com/2009/05/22/rybka-is-world-computer-champion-again-and-again/
  41. http://www.rybkachess.com/index.php?auswahl=Tournaments
  42. http://www.rybkachess.com/docs/RYBKA_EHLVEST_2007/Rybka_versus_GM_Ehlvest.htm
  43. https://journals.sagepub.com/doi/pdf/10.3233/ICG-2007-30108?download=true
  44. http://www.rybkachess.com/docs/comments_Larry_Kaufman.htm
  45. https://en.chessbase.com/post/the-dzindzi-rybka-3-handicap-match
  46. https://talkchess.com/viewtopic.php?t=22455
  47. https://www.chess.com/blog/SamCopeland/hikaru-nakamura-crushes-computer-in-blitz-chess-top-10-of-the-2000s-rybka-vs-nakamura-2008
  48. https://www.chessgames.com/perl/chessgame?gid=1497429
  49. https://www.chess.com/blog/michechess89/nakamura-crushes-rybka-in-271-moves-video
  50. https://www.youtube.com/watch?v=3bN68kfUFiA
  51. https://www.chess.com/article/view/how-rybka-and-i-tried-to-beat-the-strongest-chess-computer-in-the-world
  52. https://ssdf.bosjo.net/list.htm
  53. https://en.chessbase.com/post/rybka-fritz-lead-on-the-computer-rating-lists/4
  54. https://computerchess.org.uk/ccrl/402.archive/cgi/engine_details.cgi?print=Details&eng=Rybka%25201.0%252032-bit
  55. https://computerchess.org.uk/ccrl/402.archive/cgi/engine_details.cgi?print=Details%2B%28text%29&eng=Rybka%25202.3.2a%252064-bit
  56. https://rebel13.nl/misc/stats.html
  57. https://computerchess.org.uk/ccrl/4040/rating_list_all.html
  58. https://computerchess.org.uk/ccrl/402.archive/cgi/engine_details.cgi?print=Details%2B%28text%29&eng=Rybka%25204%252064-bit
  59. https://centaur.reading.ac.uk/80283/1/SSDF%2520List%25202018-10.pdf
  60. https://www.chess.com/news/view/chess-engine-controversy
  61. https://www.chess.com/news/view/programmers-write-open-letter-about-rybka-fruit-issue
  62. https://www.extremetech.com/extreme/88610-rybka-the-worlds-best-chess-engine-outlawed-and-disqualified
  63. https://www.chess.com/news/view/rybka-disqualified-and-banned-from-world-computer-chess-championships
  64. https://en.chessbase.com/post/icga-rybka-controversy-feedback
  65. https://scholarlypublications.universiteitleiden.nl/access/item%253A2892599/view
  66. https://www.chessprogramming.org/images/f/f5/RybkaInvestigation.pdf
  67. https://www.chess.com/news/view/rybka-affair-rajlich-speaks
  68. https://en.chessbase.com/post/feedback-on-the-icga-rybka-disqualification-scandal
  69. https://rebel13.nl/home/icga.html
  70. https://en.chessbase.com/post/10-years-houdini
  71. https://shop.chessbase.com/en/products/rybka_4
  72. https://www.chesscentral.com/pages/chess-programs-analysis/rybka-4-in-depth-report.html?srsltid=AfmBOooJPpO9bWbgjRO6_9px13rqt_ca9Bq8AqVnhuOSu3F5jRY4lO2e
  73. http://www.rybkachess.com/index.php?auswahl=FAQ
  74. https://www.chess.com/forum/view/general/rybka-still-being-sold-despite-ban-and-alleged-plagiarism
  75. http://www.rybkachess.com/index.php?auswahl=Purchase%2BRybka
  76. https://www.ebay.com/itm/127392328435
  77. http://www.rybkachess.com/index.php?auswahl=Demo%2Bversion
  78. https://en.chessbase.com/post/a-gro-miscarriage-of-justice-in-computer-che-part-two
  79. https://www.nytimes.com/2008/02/17/crosswords/chess/17chess.html
  80. https://www.reddit.com/r/chess/comments/1fejjgp/what_happened_in_2006_in_computer_chess/
  81. https://www.chess.com/forum/view/general/why-rybka-is-stronger-than-any-other-engine
  82. https://www.reddit.com/r/chess/comments/1gsq9ns/20_years_of_chess_engine_development/
Add your contribution
Related Hubs
User Avatar
No comments yet.