In the Wild West, Samuel Colt described his revolvers as “equalisers” — they gave physically weaker persons a means of defence against bullies. Engines and databases are equalisers for players from places without good chess eco-systems.
Databases launched with tiny 1000 game sets in the late 1980s. Chess engines became a serious factor from the early 90s. The gains in playing strength were incremental and mainly based on better hardware, until the early 2000s.
In the early 2000s, a new generation of engines like Fruit and Rybka developed new search-evaluation functions. By 2006, world class players could no longer match engines on standard laptops. Circa 2012, a smartphone engine can beat Super GMs.
Electronic aids also changed the thrust of research. Typically, a GM runs topical games through several engines and watches variations, seeking interesting ideas and details. Opening surprises are rarely explosively effective. People respond quickly and efficiently to novelties since their engines have also “seen” the key ideas. Preparation is now more about middlegame nuances and making the opponent subjectively uncomfortable.
Engines also play endgames of up to 7 pieces perfectly. This has in turn, led to improved technical skills. In correspondence, it means inhuman perfection. Even in normal play, the average player is less liable to endgame error.
An example of the kind of tricky endgame computers play flawlessly. WHITE TO PLAY, Study by Nasimovich 1936. This has 8 pieces so it's beyond current tablebases but the tablebases kick in once it drops below 7 pieces.
White must defend the Kt. The only way is 1. Kg3 since 1. Kg1 Bb6 2. Rh5+ Kg6 3. Rh2 Rxf2 4. Rxf2 a5 is trivially won, and so is 1. Kg2 Rxf2+. Now if 1. Kg3 Bb6 2. Nd3! unpins with a draw.
More From This Section
So black tries 1. Kg3 Rxf2! 2. Rh5+ Kg6 3. Rd5! Bb6 4. Rd6+ Kf5 5. Rxb6 Rxf3+ and now, white must play 6. Kg2!! ab6 7. Kxf3. The tempo loss leadsa to a standard draw with 7.--Ke5 8. Ke3 Kd5 9. Kd3 Kc5 10. Kc3 b5 11. Kb3 b4 12. Kb2 Kc4 13. Kc2 b3+ 14. Kb2 Kb4 15. Kb1 Kc3 16. Kc1 b2+ 17. Kb1 Kb3 stalemate.
If instead 6.Kxf3? axb6 7.Ke3 Ke5 8.Kd3 Kd5 9.Kc3 Kc5 10.Kb3 Kb5 11.Kc3 Ka4 12.Kb2 Kb4 13.Kc2 Ka3 14.Kb1 Kb3 15.Kc1 Ka2 16.Kc2 b5, black will queen and win. The pawn endgame is a standard case of King-opposition; every Russian schoolboy knows it. But under the gun, a human might make an automatic capture. A computer won't. Of course, the rook ending of 6. Kg2 Re3 7. Ra6 Re7 8. Kf3 is also easily drawn.
Devangshu Datta is an internationally rated chess and correspondence chess player