Frequently Asked Questions
The full name of Tarrasch is the Tarrasch Chess GUI. What does GUI mean ?
GUI stands for graphical user interface. The world of chess computing is divided into two parts, engines and GUIs. Engines are chess 'brains'. Engines can analyze a position and choose a move. If an engine is a brain, then a GUI is a body. The GUI provides a way for the engine to interact with the outside world. The GUI draws an on-screen chessboard so that the human user can track the current position. It allows the user to ask the engine to play a game or analyze a position. The GUI also provides a way for the user to make moves, and keeps a record of those moves.
Why is Tarrasch called Tarrasch ?
Siegbert Tarrasch was a world class player around the turn of the last century. I feel he has unfairly been stigmatized as a 'bad guy' in the historic battle of chess ideas that raged around that time. This has become almost an unthinking cliche. My program needed a name, by calling it Tarrasch I commemorate a great player's memory and perhaps in a tiny way try to rebalance this unjustice.
How does Tarrasch's opening book work ?
Tarrasch offers an unsophisticated, but useful opening book feature. Any .pgn file can be used as the book. The default opening book shipped with Tarrasch is simply a collection of a few thousand recent grandmaster games. When the book is enabled, a green onscreen indication appears whenever the current position is in the book. Green is used to indicate 'book move' in Tarrasch. Hovering the mouse over the indicator shows the book move(s), and lets you easily play one of them.
The engine uses the book, so it seemed only fair to me that a human playing against the engine should also be able to use it! Note that book moves are listed in order of popularity. If multiple book moves are available, the engine will make a random pick (biased towards more popular moves), but it will avoid a move it has played before. In other words, if you play the engine as white twice, it won't play the Sicilian both times.
Optionally Tarrasch will even ask you to reconsider your choice of move if a book move is available, but you played something else. This is the "Suggest book move" option, turned off by default. This can be a useful training feature, especially if you have created custom training material you are trying to memorize.
What engines does Tarrasch include ?
Currently standard Tarrasch ships only with the free demo version of Rybka. Although commercial Rybka versions are stronger, the demo version is perfectly satisfactory for casual chess computing. Nevertheless the engine world is rapidly changing, new and strong engines are emerging and Rybka is no longer as preeminent as it once was, so the choice of engine(s) in the standard Tarrasch package is presently under review.
Note that Komodo, a strong new engine, offers an alternative Tarrasch package on its website that includes Komodo rather than Rybka.
What is the Tarrasch Toy engine ?
Actually, there is a second engine in the standard Tarrasch package, my own Tarrasch Toy Engine. This was an engine I wrote early in Tarrasch's development as a learning exercise. I still include it in the standard package, because it can be fun to play against an engine that you can actually beat! This is particularly relevant because Tarrasch has training modes where you voluntarily handicap yourself (for example; blindfold mode). Playing a strong engine in these circumstances is no fun at all. Playing and beating Tarrasch Toy engine blindfold is a realistic and satisfying challenge for a strong player.
What is the status of the Tarrasch Toy engine ?
I haven't done any development on this for a long time. There are two versions, both stable and reliable. The older version is V0.905, and is included in the standard package. The newer and slightly stronger version is V0.906 and this is available as a separate download. I still use the older version in the standard package because it uses time less well, and so moves faster, making it a more fun opponent. You can find many Tarrasch Toy Engine games on Olivier Deville's excellent OpenWar website. Tarrasch Toy competes with reasonable success in the bottom division of the most recent tournaments.
What engines does Tarrasch support ?
In principle Tarrasch can use any UCI (a chess computing standard) engine.
How do I download and use an engine ?
New users should note that you don't have to perform this step to start using Tarrasch. Tarrasch installation requires nothing more than a few clicks of the mouse. These instructions are for people who want to install and use alternative engines.
Typically Tarrasch will already have installed engines at directories c:\Program Files (x86)\Tarrasch\Engines (Vista and Windows 7) or c:\Program Files\Tarrasch\Engines (Windows XP). So when you download an engine from the internet to your computer, it makes sense to put the new engine in the same place. An engine file should have the extension .exe. If your download has a .zip extension, you'll need to open the .zip file and copy the .exe to the desired location manually. If these instructions make no sense google "how to unzip a file" or similar to learn more. Once you have an apparently suitable .exe file in place it is worthwhile to execute it (double click it) to check it really is a UCI engine. A UCI engine will put up a nearly empty dark window and await instructions. Press "uci" then enter and if all is well the UCI engine will respond with a lot of cryptic information. Then press "quit" then enter and the dark window will disappear. If all is well choose the new engine with Options -> Engine within Tarrasch and try it out.
An important point is that most engines today come in either 32 or 64 bit forms. New PCs tend to use 64 bit Windows. These machines can run either 32 or 64 bit engines, use 64 bit engines for best performance. Older PCs can only run 32 bit engines. If you don't know whether your PC is using 32 or 64 bit Windows you can check by clicking Start -> Control Panel -> System.
Can Tarrasch run as a portable (no install) application ?
Yes. I haven't set it up for this option, but it is easy enough to do it manually. Simply copy Tarrasch.exe and an empty Tarrasch.ini file to any directory and run it from there.
Can Tarrasch run on Linux ?
Currently only using WINE. Lloyd Standish has posted instructions in my blog (link coming soon)
Can Tarrasch run on Mac ?
Currently only using WINE. Yves Catineau has posted instructions in my blog (link coming soon)
How do I make new training material ?
The old Tarrasch FAQ included a hack here that described how to create a .pgn file with Tarrasch V1. Of course Tarrasch V2 doesn't need a hack for this task which is now a smooth part of normal Tarrasch operation. So, simply create a .pgn file containing all your training material and (temporarily) set this new file as Tarrasch's book. Turn on the "Suggest book move" option, and set the book percentage to 100 (so that Tarrasch will always play a book move if it has one). You can now play games against the engine, and these games will always go down the paths established by your training material, including gentle suggestions if you deviate from the book.
Your custom training book may include games that are simply positions. This triggers a very useful Tarrasch feature; These positions can now be immediately selected from the "Training Positions" field of Tarrasch's "Setup Position" feature. Tarrasch uses whatever text appears in the "White" field associated with the position in the book file. You can see examples of this in the standard book shipped with Tarrasch, for example the text "Beginner: Mating K+Q versus bare K" selects a position which allows a beginner to practice this fundamental skill. This is a very useful feature for chess teachers. Create practice positions for your students to play out against an engine. Tarrasch automatically saves everything to a log file, so once your students have played out the positions they can send you the log and you can see how they performed!
What is a kibitzer ?
This is a Yiddish term that's found its way into chess parlance. A kibitzer is an (annoying) person who kibitzs, that is offers advice that tends to be noisy, unwanted and continual. Imagine an obnoxious older Jewish guy, probably from New York City, watching you play chess and pointing out the inadequacy of your play (in his opinion). A hint of humor is definitely implied by the use of this term in chess computing.
To start the kibitzer use the "robot" icon in the toolbar. (Or the commands menu, or the control-K key combination). The currently selected engine will start to display its evaluation of the position currently on the board. Usually the engine will show the best four moves it considers are available, in order, with follow up play. A score (in units of one pawn) will also be offered. Positive scores mean white is better, at least as far as the engine is concerned.
If the engine is calculating a move in a human versus computer game, it doesn't display the best four moves, instead it displays increasingly deep analysis of the best move it has currently found. This situation is distinguished on screen by using the term analysis rather than kibitzing.
When I stop the kibitzer, how come the kibitzing is still shown ?
This is my attempt to fix a usability problem. Often (in other chess programs) I have seen an intriguing piece of kibitzing/analysis on screen which I'd like to look at more. Typically the engine moves on to deeper analysis and overwrites the line I'm interested in before I've even memorized it. In Tarrasch in this situation, simply stop the kibitzer by pressing the robot button again. You can then look at the analysis at your leisure, or even capture it as variations using one more button click.
What is Reptor ?
Reptor was the first Triple Happy chess program. Reptor is a chess training program. It is intended to help you memorize chess knowledge, typically opening lines but also technical endgame material. Basically Reptor works by challenging the user to play down one side of a line from a database. There are other programs available that work on the same principle. I wrote Reptor because of frustrations I personally experienced with other programs. In particular I wanted a program that didn't frustrate me while I was learning the material. If I haven't yet memorized the right move, I wanted options, I don't want the training program to simply reject my guess and demand that I try again. And I particularly don't want the training program to demand I play down an arbitrarily long line, possibly into an endgame! Also, unlike the program that inspired it, Reptor is smart enough to not repeat lines I've already covered when there are lines I haven't yet played through!
I am no longer working on Reptor. Tarrasch is a more open-ended program concept, with more room to grow and improve. Nevertheless, Reptor is a complete, stable and capable program that many people may find useful. So I will continue to make it freely available here. Two versions are available, one uses Reptor's original graphics, the other uses larger, better graphics derived from Tarrasch development.
How do I run engine v engine matches with Tarrasch ?
I get asked this one a lot. Unfortunately Tarrasch doesn't have this feature yet. This feature is the number one priority for Tarrasch V3 in the future.