Since each engine has its own style of play… which engines are more positional/strategic and which engines are more sharp/tactical?

Question

I’ve heard that each engine has its own style of play. Some are more positional and strategic, while others are more sharp and tactical.

And so my question is: could you roughly sort the following engines from the most sharp and tactical to the most positional and strategic?:

• Critter 1.6a
• Deep Fritz 14
• Gull 3
• Hiarcs 14
• Houdini 4
• Komodo 7a
• Rybka 4
• Stockfish 5

And does the style of an engine depend on the time it takes to get to a high depth? I was thinking that maybe, for engines of roughly the same strength, engines that quickly get to a very high depth are more sharp and tactical, while engines that do not quickly get to a high depth have a richer and more complex evaluation function and therefore they are maybe more positional and strategic.

Answer 1

EDIT: With the advent of Lc0 and the new Stockfish 13, unfortunately this question is now beyond human comprehension.

I am not very well versed on the theory of chess computer programs, but I've used them for a long time. At first, I used Houdini, then switched for a few months to Rybka, back to Houdini, had another short incursion into Komodo, back to Houdini and now I've been using Stockfish for a year or so. The following is based on my personal feelings with the aforementioned engines (and thus, subjective).

1. Rybka: It is the model of a tactical player. Playing on (very) low depth, some times deviates from the opening theory just to see if the opponent would fall into a well-hidden trap. Playing on normal depth, it takes a little to make his mind, but he finally chooses (usually) a theoretical line. It also relies strongly on tricks and tactics to avoid calming the position (although not always with success). When in a closed position, this engine collapses.
2. Komodo: It plays half-positional, half-tactical. That is, he has no preference over either, but has good calculating power and can play relatively good in semi-closed positions. His endgame skills, however, are what differentiate him from other engines. There he is hard to beat when in an inferior position and plays accurately when having a small advantage.
3. Houdini: This is a good combination of tactical abilities and positional understanding. Can evaluate well (not great, but well) closed positions and two of his main characteristics are long combinations involving piece sacrifices to recover the piece and even win a couple of pawns after 10 to 15 moves and a tremendous defensive capability, being able to hold a draw in seemingly lost positions (it is said that this is where the name came from, in reference to his escapist abilities). Not extremely good at the endgame.
4. Stockfish: Extreme versatile engine, can calculate long and precise and has a good understanding of the positional nuances of a closed position. There are few positions in which he cannot perform well, and these can be easily tuned from the "options" available in any GUI (that is, change the way he always looks for a win or if he can allow a draw when having material advantage). However, I think it has a small tendency to drawish positions.

Answer 2

Someone on Talkchess did a test getting the top engines to play against each other in an opening which required an understanding of pawn breaks, such as the King's Indian or Stonewall. Houdini, Komodo, Rybka and Stockfish all struggled in these positions refusing to advance their pawns, and pursuing ephemeral tactical gains instead. At blitz time controls, engines overestimate the value of space and refuse to play obvious freeing moves.

Another area where engines fall down is planning for the endgame. Human grandmasters aim for certain positions which are known to be theoretically won or drawn, but engines are always caught up in short term goals, and quite willingly enter positions that are lost strategically. I think that it was Larry Kaufman who said that top engines' evaluation functions at around a 2300 or 2400 Elo level, but they never makes tactical mistakes.

Both Kasparov and Anand have commented on Hiarcs' positional understanding being more notable than other engines. Tord Romstad, one of the authors of Stockfish, acknowledged this in a fairly recent comment when he noted that on slow hardware, Hiarcs plays surprisingly well suggesting its strength lies more in its evaluation. Hiarcs has an elaborate evaluation schema, but the top engines tend to rely on a fast search.

Some engines are noted for playing in a sharp style: Tao, TheKing Tribute, Thinker and Zappa Mexico Dissident Aggressor. Some engines have parameters for sacrifices or king attack or their contempt works this way: Amyan, Fritz 10, Gambit Tiger, Junior 7 and Rybka Dynamic.