Visualize search in Fritz 10

5/21/2007 – Planning is an important ingredient of chess -- possibly the most important one. A new Fritz10 feature focuses on this aspect of the game and lets you see what Fritz is "planning" in a given position. Learn more about the "Visualize search" feature in the latest installment of ChessBase Workshop.

ChessBase 14 Download ChessBase 14 Download

Everyone uses ChessBase, from the World Champion to the amateur next door. Start your personal success story with ChessBase 14 and enjoy your chess even more!


Along with the ChessBase 14 program you can access the Live Database of 8 million games, and receive three months of free ChesssBase Account Premium membership and all of our online apps! Have a look today!

More...

Visualization is an important chess skill; anyone who's been playing the game for any length of time already knows this. The ability to accurately calculate variations is what most people think of when "visualization" is mentioned. But what this concept really means is the ability to move the pieces around in your imagination and to remember (at least temporarily) the resulting board positions.

This is an important component of planning. Many years ago I came across an article written by Jeremy Silman in which he discussed the technique (he called it an "art") of making plans. At its core, this technique consists of imagining the board the way you want it to be, in other words envisioning a board position that would be advantageous to you, and then working backwards to find a way to make that position happen. What pieces would you have to move? Which of your opponents' pieces prevent your ideal position from occurring? Can you deflect, capture, or otherwise displace those opposing pieces? Are some of your own pieces in the way? If so, what do you need to do to get them out of the way?

This sounds pretty simple, but it takes some practice to successfully add this to one's chess arsenal. There are subtechniques which one needs to master, such as visualizing a position that's achievable and which makes sense. It's not terribly beneficial to envision a back-rank mate of your opponent's King when it's still only the first or second move of the game, but such a mate might certainly be possible at some later stage of the game. In the early going, it's more beneficial to imagine where to put those two minor pieces which are still on your own first rank, the two that are preventing you from connecting your Rooks.

Silman's article hit me like a thunderbolt, and his techniques are still ones I use to this day. Basically he was talking about planning. And in many cases it really boils down to seeing in your mind what you want to have happen, and then finding a way to make that possible.

This concept is what the new "Visualize search" feature in Fritz10 is all about. Instead of a nuts-and-bolts step-by-step rehash of the best variation found so far (which you can get by using the variation board in the engine analysis pane), this new feature much more closely resembles the way a human player visualizes a plan. Let's look at a typical board position and see the feature in action:

In the above illustration we see in a graphic way, superimposed over the chessboard, what Fritz is "thinking". And this visualization process applies to both sides in the game, not just the moving player's possible future actions.

Whenever you start "infinite analysis" of a position (Engine menu/"infinite analysis"), you will see these arrows appear after a few moments. Orange arrows illustrate possible plans for White, while blue arrows show possible Black ideas. The green squares designate the preferred move for both sides in Fritz's opinion.

It's Black to move in the illustration above. Fritz thinks the best idea is ...Be6 (to be followed later by ...Bd5, as you can see from the blue arrow from e6 to d5). Other ideas under consideration are ...Bf5, which sets up a possible ...Nf6-d5-b4-c2 (assuming White plays c3-c4 as shown by an orange arrow). Another Black idea is ...Qb6, to be followed by ...Rd8 (after the Black Queen clears the d8-square).

White ideas in this position are Bf4 (Fritz' preference), as well as Ne5, Qb3 (which will cover b4 and make c3-c4 possible without the Black Knight's intrusion as described above), and h2-h3 which keeps the Black Knight off of g4.

An interesting aspect of the Visualize search feature is that the main line (preferred) variation isn't the only one illustrated; other variations, ranked a bit lower in Fritz' hierarchy of preferred courses of play, are also being shown.

It's also interesting to note that this is a dynamic feature, not a static one. The arrows displayed will change as Fritz looks deeper into a position and begins to consider other ideas. The illustration above was taken just after Fritz10 started the fourteenth ply of its search. We'll allow Fritz10 to search a bit deeper and see if the ideas change:

Now we should note that the preferred ideas for both sides have changed. Fritz still likes ...Be6 for Black but has expanded upon that idea as now sees ...Be6 as a stepping stone to ...Bd5 (there's also a possible longer sequence of ...Bf5-e6-d5). The move ...Qb6 is still under consideration, but notice that the followup ...Rd8 is absent along with the long Knight sequence which Fritz was considering earlier.

The biggest change in ideas for White is that Ne5 is now the preference. Note, too, that Fritz is considering a followup move: after the Knight vacates the square f3, the White Queen might take up a post on that spot.

For nearly ten years in this column I've been stressing ideas over variations (particularly in the opening); understanding ideas will take you farther than will the rote memorization of variations. Up until now, Fritz has strictly been a "variation machine", showing you the best line(s) of play from a given position in Infinite analysis mode. But, in my opinion, the engine has taken a huge leap forward into the realm of ideas with this new Visualize search feature. The program is making us think about ideas and doing so very simply, such as with the arrow from d1 to f3 in the last illustration. "How can the Queen go to f3? There's a Knight on that square. Wait -- there's an arrow showing the Knight moving from f3 to an advanced post on e5. So the Knight moves first and then the Queen comes to f3."

Or, returning to the first illustration above, look at what had to be considered in moving the Black Knight to c2 (forking the White Rooks): ...Nd5 might induce c3-c4 after which the Knight could go to b4 etc. Ideally this sequence would be prepared by ...Qc6 and ...Bf5, covering the Knight on its journey -- thus the ideas are to control the key squares b4 and c2, resulting in Black possibly winning the Exchange. Note that White's concrete replies aren't necessarily considered here -- this is all about an "ideal position" which Black would like to reach. Now go back and reread the first four paragraphs of this column. Isn't that quite similar to what Fritz10 is showing us here?

Some players may find this new graphic mode distracting; for these folks there's a way to switch off "Visualize search". Just go to the Tools menu, select "Options", click the "Engine" tab and uncheck "Show plans".

Note that Fritz10 is presently (as of March 2007) the only engine which utilizes this feature; it won't work with older chess engines, and at this time it remains to be seen whether or not it will be implemented in future non-Fritz ChessBase engines.

Until next week, have fun!

You can e-mail me with your comments on ChessBase Workshop. All responses will be read, and sending an e-mail to this address grants us permission to use it in a future column. No tech support questions, please.


© 2007, Steven A. Lopez. All rights reserved.



Topics f10
Discussion and Feedback Join the public discussion or submit your feedback to the editors


Discuss

Rules for reader comments

 
 

Not registered yet? Register